[R] Test (ignore)
Daniel J. Nordlund Research and Data Analysis Washington State Department of Social and Health Services Olympia, WA 98504-5204 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] test for contingency table when there are many zeros
Here is my table tt A B 1 297 398 2 470 376 3 30 23 4 3 3 5 0 0 b/c two cells are zero, I can't use chisq.test() in R which gives the following output; chisq.test(tt) Pearson's Chi-squared test data: tt X-squared = NaN, df = 4, p-value = NA Warning message: Chi-squared approximation may be incorrect in: chisq.test(tt) What function should I use then? Any suggestion? [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] test for contingency table when there are many zeros
fisher.test(tt) Francisco gallon li wrote: Here is my table tt A B 1 297 398 2 470 376 3 30 23 4 3 3 5 0 0 b/c two cells are zero, I can't use chisq.test() in R which gives the following output; chisq.test(tt) Pearson's Chi-squared test data: tt X-squared = NaN, df = 4, p-value = NA Warning message: Chi-squared approximation may be incorrect in: chisq.test(tt) What function should I use then? Any suggestion? [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Test (please ignore)
Please excuse this -- I need to test whether I can get through to R-help! (Have failed repeatedly today). Ted. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] test the significances of two regression lines
R-help, I'm trying to test the significance of two regression lines , i.e. the significance of the slopes from two samples originated from the same population. Is it correct if I fit a liner model for each sample and then test the slope signicance with 'anova'. Something like this: lm1 - lm(Y~ a1 + b1*X)# sample 1 lm2 - lm(Y~ a2 + b2*X)# sample 2 anova(lm1, lm2) Thanks in advance. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] test the significances of two regression lines
On 06-Aug-07 10:32:50, Luis Ridao Cruz wrote: R-help, I'm trying to test the significance of two regression lines , i.e. the significance of the slopes from two samples originated from the same population. Is it correct if I fit a liner model for each sample and then test the slope signicance with 'anova'. Something like this: lm1 - lm(Y~ a1 + b1*X)# sample 1 lm2 - lm(Y~ a2 + b2*X)# sample 2 anova(lm1, lm2) No, this will not work. From ?anova: Warning: The comparison between two or more models will only be valid if they are fitted to the same dataset. which is not the case in your example. One way to proceed is to merge the two datasets, and introduve a factor which identifies the dataset. For example: x1-rnorm(100) ; x2-rnorm(100) y1 - 0.2 + 0.1*x1 + 0.05*rnorm(100) y2 - 0.2 + 0.12*x2 + 0.05*rnorm(100) x - c(x1,x2) y - c(y1,y2) S - factor(c(rep(0,100),rep(1,100))) lm12 - lm(y ~ x*S) First look at the fit of y1~x1: summary(lm(y1~x1)) Coefficients: Estimate Std. Error t value Pr(|t|) (Intercept) 0.206042 0.004647 44.34 2e-16 *** x1 0.0913820.091382 0.004768 19.16 2e-16 *** Then the fit of y2~x2: summary(lm(y2~x2)) Coefficients: Estimate Std. Error t value Pr(|t|) (Intercept) 0.208216 0.005171 40.26 2e-16 *** x2 0.118840 0.005009 23.73 2e-16 *** so the estimated slopes idiffere by 0.118840 - 0.091382 = 0.027458 But what is the significance of this difference? Now: summary(lm12) Coefficients: Estimate Std. Error t value Pr(|t|) (Intercept) 0.206042 0.004923 41.852 2e-16 *** x 0.091382 0.005052 18.088 2e-16 *** S1 0.002174 0.006953 0.313 0.754926 x:S10.027457 0.006939 3.957 0.000106 *** so the x:S1 value is the same as the difference in slopes as estimated from lm1 and lm2; but now we have a standard error and a P-value for it. You can also use anova now: anova(lm12) Response: y Df Sum Sq Mean Sq F valuePr(F) x 1 2.26537 2.26537 946.2702 2.2e-16 *** S 1 0.00015 0.00015 0.0614 0.8045253 x:S 1 0.03749 0.03749 15.6599 0.0001060 *** Residuals 196 0.46922 0.00239 so you get the same P-value, though with anova() you do not see the actual estimate of the difference between the slopes. Hoping this helps, Ted. E-Mail: (Ted Harding) [EMAIL PROTECTED] Fax-to-email: +44 (0)870 094 0861 Date: 06-Aug-07 Time: 12:16:06 -- XFMail -- __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] test about distribution of data in a single population
Dear all, I would like to know how can I test which are the intervals of my data that have significant less or more counts than the other intervals. Example: Interval[1:200][200:400][400:600][600:800] ... more 900 hundred columns Count 122875 Thanks in advance, Best regards João Fadista Ph.d. student UNIVERSITY OF AARHUS Faculty of Agricultural Sciences Dept. of Genetics and Biotechnology Blichers Allé 20, P.O. BOX 50 DK-8830 Tjele Phone: +45 8999 1900 Direct: +45 8999 1900 E-mail: [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] Web: www.agrsci.org http://www.agrsci.org/ News and news media http://www.agrsci.org/navigation/nyheder_og_presse . This email may contain information that is confidential. Any use or publication of this email without written permission from Faculty of Agricultural Sciences is not allowed. If you are not the intended recipient, please notify Faculty of Agricultural Sciences immediately and delete this email. [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] test of CA axis
Dear All,
I am not a statistician, and was wondering if anyone could help me
with the following.
Greenacre, in his Correspondence Analysis in Practice (1993, p.173)
gives a method for testing the significance of an axis in CA where:
$\chi^2 = \lambda \times n$ where \lambda is the the eigenvalue for
the principal axis and n is the number of objects in the
analysis. The value for \chi^2 is then compared to a table of
critical values. The table in his book is a subset of Table 51 in
Pearson and Hartley 1976, Biometrica Tables for Statisticians vol II,
described as Percentage points of the extreme roots of
$|\text{\textbf{S}}\Sigma^{-1}-c\text{\textbf{I}}|=0$
Is there an easy way of doing this test in R? My main problem in
that Table 51 only gives values for a maximum of a p=10, \nu = 200
table and mine are regularly much bigger than that (although it would
be also nice to be able to put in the figures for lambda, n, p and
\nu and get the probability back).
Many thanks in advance, Kris Lockyear.
Dr Kris Lockyear
Institute of Archaeology
31-34 Gordon Square
London
phone: 020 7679 4568
email: [EMAIL PROTECTED]
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] test if files in current folder
runner wrote:
I want to test if the files are already in my current folder before I
download or copy from somewhere else. What's in my mind is to check if a
file is open-able in current folder. Is there a way to do this, like in
Perl:
if (open()) { do sth}?
To put it another way, how to extract all file names in a folder to an
array or list?
?files
?dir
BTW, Perl has nice -X functions, which allow file testing without explicit
opening:
http://perldoc.perl.org/functions/-X.html
--
View this message in context:
http://www.nabble.com/test-if-files-in-current-folder-tf3919587.html#a8347
Sent from the R help mailing list archive at Nabble.com.
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] test if files in current folder
On Thu, 14 Jun 2007, Vladimir Eremeev wrote:
runner wrote:
I want to test if the files are already in my current folder before I
download or copy from somewhere else. What's in my mind is to check if a
file is open-able in current folder. Is there a way to do this, like in
Perl:
if (open()) { do sth}?
To put it another way, how to extract all file names in a folder to an
array or list?
?files
?dir
BTW, Perl has nice -X functions, which allow file testing without explicit
opening:
http://perldoc.perl.org/functions/-X.html
which are based on those in 'test' and most Unix shells, as is R's
file.test() in package 'utils'.
--
Brian D. Ripley, [EMAIL PROTECTED]
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UKFax: +44 1865 272595
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] test for nested factors
Tim Bergsma wrote: Is there a conventional way to test for nested factors? I.e., if 'a' and 'b' are lists of same-length factors, does each level specified by 'a' correspond to exactly one level specified by 'b'? all( tapply(b, a, function(x) length(unique(x))==1 )) J. R. M. Hosking __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] test for nested factors
Is there a conventional way to test for nested factors? I.e., if 'a'
and 'b' are lists of same-length factors, does each level specified by
'a' correspond to exactly one level specified by 'b'?
The function below seems to suffice, but I'd be happy to know of a more
succinct solution, if it already exists.
Thanks,
Tim.
---
%nested.in% - function(x,f,...){
#coerce to list
if(!is.list(x))x-list(x)
if(!is.list(f))f-list(f)
#collapse to vector
x - tapply(x[[1]],x)
f - tapply(f[[1]],f)
#analyse
return(all(sapply(lapply(split(f,x),unique),length)==1))
}
CO2$Plant %nested.in% CO2[,c(Type,Treatment)] #TRUE
CO2$Plant %nested.in% (CO2$uptake mean(CO2$uptake)) #FALSE
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] test for nested factors
On 6/4/07, Tim Bergsma [EMAIL PROTECTED] wrote:
Is there a conventional way to test for nested factors? I.e., if 'a'
and 'b' are lists of same-length factors, does each level specified by
'a' correspond to exactly one level specified by 'b'?
The function below seems to suffice, but I'd be happy to know of a more
succinct solution, if it already exists.
How about:
%nested% - function(a, b) {
if (is.list(a)) a - do.call(interaction, c(a, drop=TRUE))
if (is.list(b)) b - do.call(interaction, c(b, drop=TRUE))
length(unique(a)) == length(unique(interaction(a, b, drop=TRUE)))
}
CO2$Plant %nested% CO2[,c(Type,Treatment)] #TRUE
CO2$Plant %nested% (CO2$uptake mean(CO2$uptake)) #FALSE
?
Hadley
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] test for nested factors
Here are two functions I wrote, 'is.nested' and 'are.crossed', that check
whether a factor is nested inside antoher one, or if both are crossed:
is.nested - function (factor1,factor2)
{
# only one positive number per line in the f1 * f2 crosstable
all(apply(table(factor1,factor2)0,1,sum) == 1)
}
are.crossed - function (factor1,factor2)
{ all(table(factor1,factor2) 0 ) }
Christophe Pallier
www.pallier.org
On 6/4/07, Tim Bergsma [EMAIL PROTECTED] wrote:
Is there a conventional way to test for nested factors? I.e., if 'a'
and 'b' are lists of same-length factors, does each level specified by
'a' correspond to exactly one level specified by 'b'?
The function below seems to suffice, but I'd be happy to know of a more
succinct solution, if it already exists.
Thanks,
Tim.
---
%nested.in% - function(x,f,...){
#coerce to list
if(!is.list(x))x-list(x)
if(!is.list(f))f-list(f)
#collapse to vector
x - tapply(x[[1]],x)
f - tapply(f[[1]],f)
#analyse
return(all(sapply(lapply(split(f,x),unique),length)==1))
}
CO2$Plant %nested.in% CO2[,c(Type,Treatment)] #TRUE
CO2$Plant %nested.in% (CO2$uptake mean(CO2$uptake)) #FALSE
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
--
Christophe Pallier (http://www.pallier.org)
[[alternative HTML version deleted]]
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[R] test to compare significant correlation increase
Hi! I am calculating correlation between two variables: 1. X versus Y 2. X versus Y(with a 3 steps lag) I would like to test if the correlation increase/decrease from 1 to 2 is significant or not. Is there any function in R to do this? any hints? Thanks for help :) David Riaño Center for Spatial Technologies and Remote Sensing (CSTARS) University of California 250-N, The Barn One Shields Avenue Davis, CA 95616-8527 USA 1-(517) 629-5499 http://www.cstars.ucdavis.edu/~driano/index.html http://www.cstars.ucdavis.edu __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] test to compare significant correlation increase
Just a guess (please correct if I'm way off on this), but maybe you could look at the difference in r betwen 1 2 and see if the confidence interval (http://davidmlane.com/hyperstat/B8544.html) for this value given your sample size includes 0. On 30-May-07, at 11:24 AM, David Riano wrote: Hi! I am calculating correlation between two variables: 1. X versus Y 2. X versus Y(with a 3 steps lag) I would like to test if the correlation increase/decrease from 1 to 2 is significant or not. Is there any function in R to do this? any hints? Thanks for help :) David Riaño Center for Spatial Technologies and Remote Sensing (CSTARS) University of California 250-N, The Barn One Shields Avenue Davis, CA 95616-8527 USA 1-(517) 629-5499 http://www.cstars.ucdavis.edu/~driano/index.html http://www.cstars.ucdavis.edu __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting- guide.html and provide commented, minimal, self-contained, reproducible code. -- Mike Lawrence Graduate Student, Department of Psychology, Dalhousie University Website: http://myweb.dal.ca/mc973993 Public calendar: http://icalx.com/public/informavore/Public The road to wisdom? Well, it's plain and simple to express: Err and err and err again, but less and less and less. - Piet Hein __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] test to compare significant correlation increase
In response to the original message, see http://finzi.psych.upenn.edu/R/Rhelp02a/archive/85753.html which describes two papers about comparing dependent correlations, which I think is a relevant question. (Try Google too. I didn't.) There is also a paper by Steiger in Psychological Bulletin, 1980, which I implemented in the following R code. (Not tested since I translated it from a BASIC script written in 1981.) r12 - as.numeric(readline(Difference r12 and r13, given r23. Input r12: )) r13 - as.numeric(readline(r13: )) r23 - as.numeric(readline(r23: )) N - as.numeric(readline(Number of Ss: )) rd - 1-r12*r12-r13*r13-r23*r23+2*r12*r13*r23 t2 - (r12-r13)*sqrt((N-1)*(1+r23)/(2*rd*(N-1)/(N-3)+((r12+r13)/2)^2*(1-r23)^3)) print(paste(t(,N-3,)=,t2, p=,pt(t2,N-3), one tailed,sep=)) print(Steiger, J.H. (1980). Tests for comparing elements of a) print(correlation matrix. Psychological Bulletin, 87, 245-251.) On 30-May-07, at 11:24 AM, David Riano wrote: Hi! I am calculating correlation between two variables: 1. X versus Y 2. X versus Y(with a 3 steps lag) I would like to test if the correlation increase/decrease from 1 to 2 is significant or not. Is there any function in R to do this? any hints? Thanks for help :) David Ria�o Center for Spatial Technologies and Remote Sensing (CSTARS) University of California 250-N, The Barn One Shields Avenue Davis, CA 95616-8527 USA 1-(517) 629-5499 http://www.cstars.ucdavis.edu/~driano/index.html http://www.cstars.ucdavis.edu __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting- guide.html and provide commented, minimal, self-contained, reproducible code. -- Mike Lawrence Graduate Student, Department of Psychology, Dalhousie University Website: http://myweb.dal.ca/mc973993 Public calendar: http://icalx.com/public/informavore/Public The road to wisdom? Well, it's plain and simple to express: Err and err and err again, but less and less and less. - Piet Hein __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Jonathan Baron, Professor of Psychology, University of Pennsylvania Home page: http://www.sas.upenn.edu/~baron __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] test deviation from a binomial distribution - lack of 50:50
Dear R-users, I have a data set where each observation consists of a number of trials (n.trials) that varies between 5 and 7, 6 being most common. Each trial can take either of two outcomes, success or failure. A dummy data set: n.trials - sample(5:7, 50, replace=T, prob=c(0.2, 0.6, 0.2)) success - rbinom(50, n.trials, p=0.5) failure - n.trials - success I know I could test for a deviation from 50:50 success:failure in one or the other direction using a glm with binomial errors. However, I suspect that in my 'real' data set the outcome 50:50 is underrepresented, not due to a skew in one particular direction, but rather that within each observation there are either many successes or many failures. Although I did not manage to create a dummy data set with these properties, which would be the proper way in R to test for a 'lack of 50:50 outcome' using the simple dummy data above as a starting point? Thanks in advance! Henrik -- __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] test ANOVA/ANCOVA
Hello everybody, I search a test for compare the k means but I have one quantitative variable and two groups date, traitement. And I suppose my samples are dependant with the date. What the statistical test would I use? Thank you. my data: dateEU DW wk1 EU1 5,324547829 wk1 EU1 7,321253265 wk1 EU1 4,431712065 wk1 EU2 8,230322407 wk1 EU2 8,546873269 wk1 EU2 5,657332069 wk1 EU3 3,165508618 wk1 EU3 4,431712065 wk1 EU3 1,899305171 wk2 EU1 2,163097556 wk2 EU1 17,61379438 wk2 EU1 15,82754309 wk2 EU2 16,46064481 wk2 EU2 19,30960257 wk2 EU2 13,92823792 wk2 EU3 6,014466374 wk2 EU3 7,280669822 wk2 EU3 5,064813789 wk4 EU1 11,03179753 wk4 EU1 29,75578101 wk4 EU1 22,71252433 wk4 EU2 27,85647584 wk4 EU2 36,71989997 wk4 EU2 20,11680727 wk4 EU3 13,59661321 wk4 EU3 13,2951362 wk4 EU3 14,56133964 wk6 EU1 30,73875474 wk6 EU1 33,27842393 wk6 EU1 35,27512937 wk6 EU2 31,2817185 wk6 EU2 41,53147307 wk6 EU2 35,57093758 wk6 EU3 8,652390223 wk6 EU3 18,6359174 wk6 EU3 15,30807501 _ Avec Windows Live OneCare éliminez tous les virus de votre PC ! __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] test for bimodality
Hi, I found this post from 6 years ago about this: https://stat.ethz.ch/pipermail/r-help/2001-May/012800.html Basically, I'd like to find an R implementation of the multimodality test in the book Introduction to the Bootstrap by Efron and Tibshirani. There is an R package bootstrap which is the package to go with the book, but it does not seem to implement the multimodality test. I found this post to the S mailing list giving an implementation in development: http://www.biostat.wustl.edu/archives/html/s-news/2000-06/msg00038.html but I'm not sure about it. Anyway, if anyone knows where I can find an R implementation please let me and the list know. thanks, Andrew Smith __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Test of Presence Matrix HOWTO?
Hello, Imagine 3 lists like so: a - list(A,B,C,D) b - list(A,B,E,F) c - list(A,C,E,G) What I need (vennDiagram) is a matrix characterizing with 1 or 0 whether any given member is present or not like so: x1 x2 x3 [1,] 1 1 1 [2,] 1 1 0 [3,] 1 0 1 [4,] 1 0 0 [5,] 0 1 1 [6,] 0 1 0 [7,] 0 0 1 (where the rows represent A-G and the columns a-c, respectively). table(c(a,b,c)) will give me a quick answer for the 1 1 1 case, but how to deal with the other cases efficiently without looping over each string and looking for membership %in% each list? Thanks for enlightening the learning, Joh __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test of Presence Matrix HOWTO?
a - c(A,B,C,D) b - c(A,B,E,F) c - c(A,C,E,G) Df - cbind(a, b, c) apply(Df, 2, function(x)(LETTERS[1:7] %in% x)) a b c [1,] TRUE TRUE TRUE [2,] TRUE TRUE FALSE [3,] TRUE FALSE TRUE [4,] TRUE FALSE FALSE [5,] FALSE TRUE TRUE [6,] FALSE TRUE FALSE [7,] FALSE FALSE TRUE apply(Df, 2, function(x)(as.numeric(LETTERS[1:7] %in% x))) a b c [1,] 1 1 1 [2,] 1 1 0 [3,] 1 0 1 [4,] 1 0 0 [5,] 0 1 1 [6,] 0 1 0 [7,] 0 0 1 Cheers, Thierry ir. Thierry Onkelinx Instituut voor natuur- en bosonderzoek / Reseach Institute for Nature and Forest Cel biometrie, methodologie en kwaliteitszorg / Section biometrics, methodology and quality assurance Gaverstraat 4 9500 Geraardsbergen Belgium tel. + 32 54/436 185 [EMAIL PROTECTED] www.inbo.be Do not put your faith in what statistics say until you have carefully considered what they do not say. ~William W. Watt A statistical analysis, properly conducted, is a delicate dissection of uncertainties, a surgery of suppositions. ~M.J.Moroney -Oorspronkelijk bericht- Van: [EMAIL PROTECTED] [mailto:r-help- [EMAIL PROTECTED] Namens Johannes Graumann Verzonden: maandag 26 februari 2007 16:25 Aan: r-help@stat.math.ethz.ch Onderwerp: [R] Test of Presence Matrix HOWTO? Hello, Imagine 3 lists like so: a - list(A,B,C,D) b - list(A,B,E,F) c - list(A,C,E,G) What I need (vennDiagram) is a matrix characterizing with 1 or 0 whether any given member is present or not like so: x1 x2 x3 [1,] 1 1 1 [2,] 1 1 0 [3,] 1 0 1 [4,] 1 0 0 [5,] 0 1 1 [6,] 0 1 0 [7,] 0 0 1 (where the rows represent A-G and the columns a-c, respectively). table(c(a,b,c)) will give me a quick answer for the 1 1 1 case, but how to deal with the other cases efficiently without looping over each string and looking for membership %in% each list? Thanks for enlightening the learning, Joh __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting- guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test of Presence Matrix HOWTO?
you can use something like the following: a - list(A,B,C,D) b - list(A,B,E,F) c - list(A,C,E,G) # abc - list(a, b, c) unq.abc - unique(unlist(abc)) out.lis - lapply(abc, %in%, x = unq.abc) out.lis lapply(out.lis, as.numeric) I hope it helps. Best, Dimitris Dimitris Rizopoulos Ph.D. Student Biostatistical Centre School of Public Health Catholic University of Leuven Address: Kapucijnenvoer 35, Leuven, Belgium Tel: +32/(0)16/336899 Fax: +32/(0)16/337015 Web: http://med.kuleuven.be/biostat/ http://www.student.kuleuven.be/~m0390867/dimitris.htm - Original Message - From: Johannes Graumann [EMAIL PROTECTED] To: r-help@stat.math.ethz.ch Sent: Monday, February 26, 2007 4:25 PM Subject: [R] Test of Presence Matrix HOWTO? Hello, Imagine 3 lists like so: a - list(A,B,C,D) b - list(A,B,E,F) c - list(A,C,E,G) What I need (vennDiagram) is a matrix characterizing with 1 or 0 whether any given member is present or not like so: x1 x2 x3 [1,] 1 1 1 [2,] 1 1 0 [3,] 1 0 1 [4,] 1 0 0 [5,] 0 1 1 [6,] 0 1 0 [7,] 0 0 1 (where the rows represent A-G and the columns a-c, respectively). table(c(a,b,c)) will give me a quick answer for the 1 1 1 case, but how to deal with the other cases efficiently without looping over each string and looking for membership %in% each list? Thanks for enlightening the learning, Joh __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Test for Kurtosis.
Dear all R users, I question is on the subject of Statistics rather R related. I am searching some documentation on statistical significance test for Kurtosis especially Glyn test. More specifically I am looking for documentation on: 1: Test statistic for Glyn test 2. Statistical properties of this test like distribution etc. Can anyone give me any reference on this? Any help will be highly appreciated. Thanks and regard, megh. - [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] test of spatial dependence?? - ask an ecologist?
Although you don't have coordinates, there has to be some kind of information on spatial relationships. (I am likely saying the same thing as Roger.) Let me see if I have this right. Some large area is partitioned into cells (polygons in an ArcInfo shape file?) and, within each cell you compute something. Then the distance between any two cells may be zero or one, one for any pair of cells that are adjacent, zero for any pair not adjacent. It doesn't strike me as that odd a situation and so I think I would like to know what is a conventional solution. One possibility would be to use Euclidean distances among centroids of cells. I have not reviewed the basis of the Mantel tests used by ecologists. If that is what you need, in looking for a program we would need to be mindful of the fact that a probably different Mantel test is used routinely for evaluation of disease data, in particular specialized as the logrank test of survival analysis. regards, Farrar Roger Bivand [EMAIL PROTECTED] wrote: On Wed, 6 Dec 2006, Xu Yuan wrote: Thanks David and Milton for replies. No, I don't have the coordiates. In other words, my data are not point data. But I think there is a way to test of spatial dependence for areal data or lattice data. In this case, the variable of interest is typically the average value of an area instead of a point. Do you how to do this? But do you know where the areas are in relation to each other? Does the spatialCovariance package help? Roger Thank you. Xu On 12/6/06, Milton Cezar Ribeiro wrote: I never used it, but I beleave that it is a job for mantel.rtest() available on ade4 package. In fact Farrar are right, you will neet the XY coordinates. Give a look at Legendre Legendre text book. HTH, Miltinho Brazil David Farrar escreveu: In addition to the 25 numbers, I assume you have coordinates of each field. Otherwise, I don't understand what you are trying to do. I think ecologists like to use a test due to Mantel in this situation. The prefix auto means self, of course, the idea being that measurements of the same variable under different conditions are correlated. I guess this would be a case of autodependence. For correlation versus dependence, check your intro stats book. de nada, X'X Farrar Xu Yuan wrote: hello R-friends, I am a R beginner and try to ask a basic question: How to test the spatial dependence of a column of data? for example, I have 25 agricultural fields, and I measure the average slope (%) or pH for each field. All I have is 25 numbers. PS, could someone confirm that spatial dependence is equivalent to spatial correlation or spatial autocorrelation or not. Thank you very much. XY [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. O Yahoo! está de cara nova. Venha conferir! __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Roger Bivand Economic Geography Section, Department of Economics, Norwegian School of Economics and Business Administration, Helleveien 30, N-5045 Bergen, Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43 e-mail: [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] test of spatial dependence??
How to test the spatial dependence of a column of data? for example [...] All I have is 25 numbers. If you don't have coordinates of your data, as I understand here, there's nothing you can do, of course ... If you have coordinates, you should compute the variogram -and/or the spatial covariance- of your data and look if they are meaningful. Hint: read a book about geostatistics. PS, could someone confirm that spatial dependence is equivalent to spatial correlation or spatial autocorrelation or not. So, to me there is a light difference, but it's really for advanced users, and maybe questionable. Bye. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] test of spatial dependence?? - ask an ecologist?
I never used it, but I beleave that it is a job for mantel.rtest() available on ade4 package. In fact Farrar are right, you will neet the XY coordinates. Give a look at Legendre Legendre text book. HTH, Miltinho Brazil David Farrar [EMAIL PROTECTED] escreveu: In addition to the 25 numbers, I assume you have coordinates of each field. Otherwise, I don't understand what you are trying to do. I think ecologists like to use a test due to Mantel in this situation. The prefix auto means self, of course, the idea being that measurements of the same variable under different conditions are correlated. I guess this would be a case of autodependence. For correlation versus dependence, check your intro stats book. de nada, X'X Farrar Xu Yuan wrote: hello R-friends, I am a R beginner and try to ask a basic question: How to test the spatial dependence of a column of data? for example, I have 25 agricultural fields, and I measure the average slope (%) or pH for each field. All I have is 25 numbers. PS, could someone confirm that spatial dependence is equivalent to spatial correlation or spatial autocorrelation or not. Thank you very much. XY [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. - [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] test of spatial dependence?? - ask an ecologist?
match-making is such fun. Farrar Milton Cezar Ribeiro [EMAIL PROTECTED] wrote: I never used it, but I beleave that it is a job for mantel.rtest() available on ade4 package. In fact Farrar are right, you will neet the XY coordinates. Give a look at Legendre Legendre text book. HTH, Miltinho Brazil David Farrar [EMAIL PROTECTED] escreveu: In addition to the 25 numbers, I assume you have coordinates of each field. Otherwise, I don't understand what you are trying to do. I think ecologists like to use a test due to Mantel in this situation. The prefix auto means self, of course, the idea being that measurements of the same variable under different conditions are correlated. I guess this would be a case of autodependence. For correlation versus dependence, check your intro stats book. de nada, X'X Farrar Xu Yuan wrote: hello R-friends, I am a R beginner and try to ask a basic question: How to test the spatial dependence of a column of data? for example, I have 25 agricultural fields, and I measure the average slope (%) or pH for each field. All I have is 25 numbers. PS, could someone confirm that spatial dependence is equivalent to spatial correlation or spatial autocorrelation or not. Thank you very much. XY [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. - O Yahoo! está de cara nova. Venha conferir! [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] test of spatial dependence?? - ask an ecologist?
Thanks David and Milton for replies. No, I don't have the coordiates. In other words, my data are not point data. But I think there is a way to test of spatial dependence for areal data or lattice data. In this case, the variable of interest is typically the average value of an area instead of a point. Do you how to do this? Thank you. Xu On 12/6/06, Milton Cezar Ribeiro [EMAIL PROTECTED] wrote: I never used it, but I beleave that it is a job for mantel.rtest() available on ade4 package. In fact Farrar are right, you will neet the XY coordinates. Give a look at Legendre Legendre text book. HTH, Miltinho Brazil David Farrar [EMAIL PROTECTED] escreveu: In addition to the 25 numbers, I assume you have coordinates of each field. Otherwise, I don't understand what you are trying to do. I think ecologists like to use a test due to Mantel in this situation. The prefix auto means self, of course, the idea being that measurements of the same variable under different conditions are correlated. I guess this would be a case of autodependence. For correlation versus dependence, check your intro stats book. de nada, X'X Farrar Xu Yuan wrote: hello R-friends, I am a R beginner and try to ask a basic question: How to test the spatial dependence of a column of data? for example, I have 25 agricultural fields, and I measure the average slope (%) or pH for each field. All I have is 25 numbers. PS, could someone confirm that spatial dependence is equivalent to spatial correlation or spatial autocorrelation or not. Thank you very much. XY [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. O Yahoo! está de cara nova. Venha conferir! __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] test of spatial dependence?? - ask an ecologist?
On Wed, 6 Dec 2006, Xu Yuan wrote: Thanks David and Milton for replies. No, I don't have the coordiates. In other words, my data are not point data. But I think there is a way to test of spatial dependence for areal data or lattice data. In this case, the variable of interest is typically the average value of an area instead of a point. Do you how to do this? But do you know where the areas are in relation to each other? Does the spatialCovariance package help? Roger Thank you. Xu On 12/6/06, Milton Cezar Ribeiro [EMAIL PROTECTED] wrote: I never used it, but I beleave that it is a job for mantel.rtest() available on ade4 package. In fact Farrar are right, you will neet the XY coordinates. Give a look at Legendre Legendre text book. HTH, Miltinho Brazil David Farrar [EMAIL PROTECTED] escreveu: In addition to the 25 numbers, I assume you have coordinates of each field. Otherwise, I don't understand what you are trying to do. I think ecologists like to use a test due to Mantel in this situation. The prefix auto means self, of course, the idea being that measurements of the same variable under different conditions are correlated. I guess this would be a case of autodependence. For correlation versus dependence, check your intro stats book. de nada, X'X Farrar Xu Yuan wrote: hello R-friends, I am a R beginner and try to ask a basic question: How to test the spatial dependence of a column of data? for example, I have 25 agricultural fields, and I measure the average slope (%) or pH for each field. All I have is 25 numbers. PS, could someone confirm that spatial dependence is equivalent to spatial correlation or spatial autocorrelation or not. Thank you very much. XY [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. O Yahoo! está de cara nova. Venha conferir! __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- Roger Bivand Economic Geography Section, Department of Economics, Norwegian School of Economics and Business Administration, Helleveien 30, N-5045 Bergen, Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43 e-mail: [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] test of spatial dependence??
hello R-friends, I am a R beginner and try to ask a basic question: How to test the spatial dependence of a column of data? for example, I have 25 agricultural fields, and I measure the average slope (%) or pH for each field. All I have is 25 numbers. PS, could someone confirm that spatial dependence is equivalent to spatial correlation or spatial autocorrelation or not. Thank you very much. XY [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] test of spatial dependence?? - ask an ecologist?
In addition to the 25 numbers, I assume you have coordinates of each field. Otherwise, I don't understand what you are trying to do. I think ecologists like to use a test due to Mantel in this situation. The prefix auto means self, of course, the idea being that measurements of the same variable under different conditions are correlated. I guess this would be a case of autodependence. For correlation versus dependence, check your intro stats book. de nada, X'X Farrar Xu Yuan [EMAIL PROTECTED] wrote: hello R-friends, I am a R beginner and try to ask a basic question: How to test the spatial dependence of a column of data? for example, I have 25 agricultural fields, and I measure the average slope (%) or pH for each field. All I have is 25 numbers. PS, could someone confirm that spatial dependence is equivalent to spatial correlation or spatial autocorrelation or not. Thank you very much. XY [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
Gregor Gorjanc a écrit : If I summarize the thread there is (currently) no way to test for internet presence with a general approach. what about try(readLines(...)) ? (at least it works fine on Windows.) __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
On 11-Sep-06 Gregor Gorjanc wrote: Thank you. Your solution is usable but unfortunatelly not portable to Windows. I would like to use this test in package check, which can include also windows OS. Now that I think about it (should have done that earlier), you can use the value of .Platform$OS.type (either unix or windows) to determine the OS type, and also the value of the system command to see what is sent back from the ping command. Then the R command to ping a remote host once would be system(ping -c 1 remote host) for unix system(ping -n 1 remote host) for windows (at any rate for the DOS-based ping in Windows 98). However, the returned value of system (which consists of the output of the ping commmand) is a multi-line value, and is different between Unix and Windows So to distinguish between success and failure, you would have to do some slightly complex parsing of the output returned by system, and evaluate it differently according to operating system. And the latter may vary between different versions of the OS ... Maybe it wasn't such a good suggestion after all. However, since separate versions of your package would be compiled for Unix-like systems and for Windows, perhaps a way round such differences is to write a different script for each one (shell script as before for Unix, batch file for Windows), crafted so as to return identical results in the two cases for success, and for failure. Best wishes, Ted. E-Mail: (Ted Harding) [EMAIL PROTECTED] Fax-to-email: +44 (0)870 094 0861 Date: 12-Sep-06 Time: 01:28:15 -- XFMail -- __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
Prof Brian Ripley ripley at stats.ox.ac.uk writes:
...
Check out tests/internet.R. nsl() checks if you can resolve host names,
which has worked well enough there.
Thank you prof. Ripley for this pointer. I am posting here the relevant part if
someone does not look at SVN. I would just like to ask why is .Platform$OS.type
== unix added to the test? Is nsl() available only on unix like platforms or
... I did not found any specifics in its help page.
if(!capabilities()[http/ftp]) {
warning(no internet capabilities)
q()
}
if(.Platform$OS.type == unix
is.null(nsl(cran.r-project.org))) q()
Does it make any sense to write a function that would use these two tests.
isNetAvailable - function()
{
ifelse(!capabilities()[http/ftp]
## .Platform$OS.type == unix ## ???
is.null(nsl(cran.r-project.org)),
FALSE,
TRUE)
}
Regards, Gregor
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
Gabor Grothendieck wrote: Here is a variation for Windows. The second line returns TRUE or FALSE and may need to be varied if the output of ping is not the same on your system as on mine: ping - system(ping www.google.com, intern = TRUE) as.numeric(strsplit(grep(Received, ping, value = TRUE), [ ,])[[1]][8]) 0 ... Thank you Gabor and Ted! These are all fine ways, but as both of you have stated not really general as you never know what ping will produce on different versions etc. I am really keen on test from tests/internet.R. If that test is OK for base R, I do not see why it should not be OK for R package. I just hope to get some more feedback on my question[1] to prof. Ripley. [1]https://stat.ethz.ch/pipermail/r-help/2006-September/112766.html Thank you! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
On Mon, 11 Sep 2006, Gregor Gorjanc wrote:
Prof Brian Ripley ripley at stats.ox.ac.uk writes:
...
Check out tests/internet.R. nsl() checks if you can resolve host names,
which has worked well enough there.
Thank you prof. Ripley for this pointer. I am posting here the relevant part
if
someone does not look at SVN. I would just like to ask why is
.Platform$OS.type
== unix added to the test? Is nsl() available only on unix like platforms or
... I did not found any specifics in its help page.
Did you look at the help page on Windows? Looking at the help page on
Unix only tells you about Unix.
Hint: the help page is src/library/utils/man/unix/nsl.Rd
(In my country, PhD students are supposed to be able to find things
like that out for themselves.)
if(!capabilities()[http/ftp]) {
warning(no internet capabilities)
q()
}
if(.Platform$OS.type == unix
is.null(nsl(cran.r-project.org))) q()
Does it make any sense to write a function that would use these two tests.
isNetAvailable - function()
{
ifelse(!capabilities()[http/ftp]
## .Platform$OS.type == unix ## ???
is.null(nsl(cran.r-project.org)),
FALSE,
TRUE)
}
Regards, Gregor
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
--
Brian D. Ripley, [EMAIL PROTECTED]
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UKFax: +44 1865 272595
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
Prof Brian Ripley wrote: On Mon, 11 Sep 2006, Gregor Gorjanc wrote: Prof Brian Ripley ripley at stats.ox.ac.uk writes: ... Check out tests/internet.R. nsl() checks if you can resolve host names, which has worked well enough there. Thank you prof. Ripley for this pointer. I am posting here the relevant part if someone does not look at SVN. I would just like to ask why is .Platform$OS.type == unix added to the test? Is nsl() available only on unix like platforms or ... I did not found any specifics in its help page. Did you look at the help page on Windows? Looking at the help page on Unix only tells you about Unix. Hint: the help page is src/library/utils/man/unix/nsl.Rd (In my country, PhD students are supposed to be able to find things like that out for themselves.) ... Thank you for additional pointer. I did not look windows help page as I do not have windows at disposal all the time, but You are right - I should have looked into the sources. I found out that there is no such function i.e. nsl() under windows. If I summarize the thread there is (currently) no way to test for internet presence with a general approach. -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Test internet presence
Hello useRs! I have a small package and I need internet access for examples. Of course it works fine when I have internet access, but not otherwise. I remember I saw a way to test availability of internet access, but I can not find it now in archives. Can anyone here help me? Thanks! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
David Barron wrote: I means non-zero, not non-negative. ... Thanks. download.file() is fine! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
The function download.file (invisibly) returns a non-negative integer if it fails, so I suppose you could use this as a test. On 11/09/06, Gregor Gorjanc [EMAIL PROTECTED] wrote: Hello useRs! I have a small package and I need internet access for examples. Of course it works fine when I have internet access, but not otherwise. I remember I saw a way to test availability of internet access, but I can not find it now in archives. Can anyone here help me? Thanks! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- = David Barron Said Business School University of Oxford Park End Street Oxford OX1 1HP __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
I means non-zero, not non-negative. On 11/09/06, David Barron [EMAIL PROTECTED] wrote: The function download.file (invisibly) returns a non-negative integer if it fails, so I suppose you could use this as a test. On 11/09/06, Gregor Gorjanc [EMAIL PROTECTED] wrote: Hello useRs! I have a small package and I need internet access for examples. Of course it works fine when I have internet access, but not otherwise. I remember I saw a way to test availability of internet access, but I can not find it now in archives. Can anyone here help me? Thanks! -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. -- = David Barron Said Business School University of Oxford Park End Street Oxford OX1 1HP -- = David Barron Said Business School University of Oxford Park End Street Oxford OX1 1HP __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
Gregor Gorjanc wrote: Hello useRs! I have a small package and I need internet access for examples. Of course it works fine when I have internet access, but not otherwise. I remember I saw a way to test availability of internet access, but I can not find it now in archives. Can anyone here help me? David Barron wrote: I means non-zero, not non-negative. ... Thanks. download.file() is fine! Thinking a bit more, download.file() is not really general as I need to specify an url and it might happen that I have internet connection, but server that provides url is down. I am to picky on this? -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
Gregor Gorjanc wrote:
Gregor Gorjanc wrote:
Hello useRs!
I have a small package and I need internet access for examples. Of
course it works fine when I have internet access, but not otherwise. I
remember I saw a way to test availability of internet access, but I can
not find it now in archives. Can anyone here help me?
David Barron wrote:
I means non-zero, not non-negative.
...
Thanks. download.file() is fine!
Thinking a bit more, download.file() is not really general as I need to
specify an url and it might happen that I have internet connection, but
server that provides url is down. I am to picky on this?
Additionally, download.file throws an error if url is not available i.e.
if I do not have internet access. I have used the following example:
if(download.file(url=http://www.r-project.org/index.html;,
destfile=tempfile()) == 0) {
print(Yuhuu)
}
unpluged the net cable and got the following message during R CMD check
(in examples part)
trying URL 'http://www.r-project.org/index.html'
Warning: unable to resolve 'www.r-project.org'.
Error in download.file(url = http://www.r-project.org/index.html;,
destfile = tempfile()) :
cannot open URL 'http://www.r-project.org/index.html'
Execution halted
It seems that 'internal' method was used (I use R 2.3.1 under Linux) as
indicated in help page of download.file. I could use wget or lynx
methods, but these two must be available, so this is not really
portable. Are there any other options for testing internet access? I am
thinking that this might be more relevant for R-devel. I will wait a bit
before moving there.
--
Lep pozdrav / With regards,
Gregor Gorjanc
--
University of Ljubljana PhD student
Biotechnical Faculty
Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan
Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si
SI-1230 Domzale tel: +386 (0)1 72 17 861
Slovenia, Europefax: +386 (0)1 72 17 888
--
One must learn by doing the thing; for though you think you know it,
you have no certainty until you try. Sophocles ~ 450 B.C.
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
On 11-Sep-06 Gregor Gorjanc wrote:
It seems that 'internal' method was used (I use R 2.3.1 under
Linux) as indicated in help page of download.file. I could
use wget or lynx methods, but these two must be available,
so this is not really portable. Are there any other options
for testing internet access? I am thinking that this might be
more relevant for R-devel. I will wait a bit
before moving there.
--
Lep pozdrav / With regards,
Gregor Gorjanc
Hi Gorjanc,
Since you are using Linux, I think you should ask R to delegate
the test to the system.
If you have a script, in executable file (755) say test.inet.sh,
which says something like
if ping -c 1 something ; then
export NET_UP=YES
else
export NETP_UP=NO
fi
where something is the IP address or name of an external host
which responds to 'ping' (some will not, depending on their firewall
settings), then you can use on R:
system(test.inet)
if( sys.getenv(NET_UP) == YES ) { ... } else { ... }
For example (nothing to do with R, but shows the principle),
I have the following script to set my system time and hardware
clock from whichever one of 3 NTP servers is willing to respond:
if /bin/ping -c 1 ntp0.zen.co.uk ; then
export NETTIME=/usr/sbin/ntpdate -u ntp0.zen.co.uk
elif /bin/ping -c 1 ntp2b.mcc.ac.uk ; then
export NETTIME=/usr/sbin/ntpdate -u ntp2b.mcc.ac.uk
elif /bin/ping -c 1 ntp2c.mcc.ac.uk ; then
export NETTIME=/usr/sbin/ntpdate -u ntp2c.mcc.ac.uk
else
export NETTIME=
fi
if [ $NETTIME != ] ; then
sleep 1
sleep 1
$NETTIME
/sbin/clock -u -w
date
fi
which also illustrates how to allow for the possibility that
the default server might not be responding at the time, so
it has 2 fallback servers.
Hoping this helps,
Ted.
E-Mail: (Ted Harding) [EMAIL PROTECTED]
Fax-to-email: +44 (0)870 094 0861
Date: 11-Sep-06 Time: 13:42:22
-- XFMail --
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
On Mon, 11 Sep 2006, [EMAIL PROTECTED] wrote: On 11-Sep-06 Gregor Gorjanc wrote: It seems that 'internal' method was used (I use R 2.3.1 under Linux) as indicated in help page of download.file. I could use wget or lynx methods, but these two must be available, so this is not really portable. Are there any other options for testing internet access? I am thinking that this might be more relevant for R-devel. I will wait a bit before moving there. Check out tests/internet.R. nsl() checks if you can resolve host names, which has worked well enough there. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
[Sorry -- errors due to mistyping especially at below.
Corrected in the following lines.]
On 11-Sep-06 Ted Harding wrote:
On 11-Sep-06 Gregor Gorjanc wrote:
It seems that 'internal' method was used (I use R 2.3.1 under
Linux) as indicated in help page of download.file. I could
use wget or lynx methods, but these two must be available,
so this is not really portable. Are there any other options
for testing internet access? I am thinking that this might be
more relevant for R-devel. I will wait a bit
before moving there.
--
Lep pozdrav / With regards,
Gregor Gorjanc
Hi Gregor,
Since you are using Linux, I think you should ask R to delegate
the test to the system.
If you have a script, in executable file (755) say test.inet.sh,
which says something like
if ping -c 1 something ; then
export NET_UP=YES
else
export NETP_UP=NO
fi
where something is the IP address or name of an external host
which responds to 'ping' (some will not, depending on their firewall
settings), then you can use on R:
system(test.inet)
if( sys.getenv(NET_UP) == YES ) { ... } else { ... }
system(test.inet.sh)
if( sys.getenv(NET_UP) == YES ) { ... } else { ... }
For example (nothing to do with R, but shows the principle),
I have the following script to set my system time and hardware
clock from whichever one of 3 NTP servers is willing to respond:
if /bin/ping -c 1 ntp0.zen.co.uk ; then
export NETTIME=/usr/sbin/ntpdate -u ntp0.zen.co.uk
elif /bin/ping -c 1 ntp2b.mcc.ac.uk ; then
export NETTIME=/usr/sbin/ntpdate -u ntp2b.mcc.ac.uk
elif /bin/ping -c 1 ntp2c.mcc.ac.uk ; then
export NETTIME=/usr/sbin/ntpdate -u ntp2c.mcc.ac.uk
else
export NETTIME=
fi
if [ $NETTIME != ] ; then
sleep 1
sleep 1
$NETTIME
/sbin/clock -u -w
date
fi
which also illustrates how to allow for the possibility that
the default server might not be responding at the time, so
it has 2 fallback servers.
Hoping this helps,
Ted.
E-Mail: (Ted Harding) [EMAIL PROTECTED]
Fax-to-email: +44 (0)870 094 0861
Date: 11-Sep-06 Time: 13:42:22
-- XFMail --
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
E-Mail: (Ted Harding) [EMAIL PROTECTED]
Fax-to-email: +44 (0)870 094 0861
Date: 11-Sep-06 Time: 14:28:11
-- XFMail --
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
Thank you. Your solution is usable but unfortunatelly not portable to
Windows. I would like to use this test in package check, which can
include also windows OS.
On 11-Sep-06 Ted Harding wrote:
On 11-Sep-06 Gregor Gorjanc wrote:
It seems that 'internal' method was used (I use R 2.3.1 under
Linux) as indicated in help page of download.file. I could
use wget or lynx methods, but these two must be available,
so this is not really portable. Are there any other options
for testing internet access? I am thinking that this might be
more relevant for R-devel. I will wait a bit
before moving there.
--
Lep pozdrav / With regards,
Gregor Gorjanc
Hi Gregor,
Since you are using Linux, I think you should ask R to delegate
the test to the system.
If you have a script, in executable file (755) say test.inet.sh,
which says something like
if ping -c 1 something ; then
export NET_UP=YES
else
export NETP_UP=NO
fi
where something is the IP address or name of an external host
which responds to 'ping' (some will not, depending on their firewall
settings), then you can use on R:
system(test.inet)
if( sys.getenv(NET_UP) == YES ) { ... } else { ... }
system(test.inet.sh)
if( sys.getenv(NET_UP) == YES ) { ... } else { ... }
For example (nothing to do with R, but shows the principle),
I have the following script to set my system time and hardware
clock from whichever one of 3 NTP servers is willing to respond:
if /bin/ping -c 1 ntp0.zen.co.uk ; then
export NETTIME=/usr/sbin/ntpdate -u ntp0.zen.co.uk
elif /bin/ping -c 1 ntp2b.mcc.ac.uk ; then
export NETTIME=/usr/sbin/ntpdate -u ntp2b.mcc.ac.uk
elif /bin/ping -c 1 ntp2c.mcc.ac.uk ; then
export NETTIME=/usr/sbin/ntpdate -u ntp2c.mcc.ac.uk
else
export NETTIME=
fi
if [ $NETTIME != ] ; then
sleep 1
sleep 1
$NETTIME
/sbin/clock -u -w
date
fi
which also illustrates how to allow for the possibility that
the default server might not be responding at the time, so
it has 2 fallback servers.
--
Lep pozdrav / With regards,
Gregor Gorjanc
--
University of Ljubljana PhD student
Biotechnical Faculty
Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan
Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si
SI-1230 Domzale tel: +386 (0)1 72 17 861
Slovenia, Europefax: +386 (0)1 72 17 888
--
One must learn by doing the thing; for though you think you know it,
you have no certainty until you try. Sophocles ~ 450 B.C.
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test internet presence
Here is a variation for Windows. The second line returns TRUE or FALSE
and may need to be varied if the output of ping is not the same on your
system as on mine:
ping - system(ping www.google.com, intern = TRUE)
as.numeric(strsplit(grep(Received, ping, value = TRUE), [ ,])[[1]][8]) 0
On 9/11/06, Gregor Gorjanc [EMAIL PROTECTED] wrote:
Thank you. Your solution is usable but unfortunatelly not portable to
Windows. I would like to use this test in package check, which can
include also windows OS.
On 11-Sep-06 Ted Harding wrote:
On 11-Sep-06 Gregor Gorjanc wrote:
It seems that 'internal' method was used (I use R 2.3.1 under
Linux) as indicated in help page of download.file. I could
use wget or lynx methods, but these two must be available,
so this is not really portable. Are there any other options
for testing internet access? I am thinking that this might be
more relevant for R-devel. I will wait a bit
before moving there.
--
Lep pozdrav / With regards,
Gregor Gorjanc
Hi Gregor,
Since you are using Linux, I think you should ask R to delegate
the test to the system.
If you have a script, in executable file (755) say test.inet.sh,
which says something like
if ping -c 1 something ; then
export NET_UP=YES
else
export NETP_UP=NO
fi
where something is the IP address or name of an external host
which responds to 'ping' (some will not, depending on their firewall
settings), then you can use on R:
system(test.inet)
if( sys.getenv(NET_UP) == YES ) { ... } else { ... }
system(test.inet.sh)
if( sys.getenv(NET_UP) == YES ) { ... } else { ... }
For example (nothing to do with R, but shows the principle),
I have the following script to set my system time and hardware
clock from whichever one of 3 NTP servers is willing to respond:
if /bin/ping -c 1 ntp0.zen.co.uk ; then
export NETTIME=/usr/sbin/ntpdate -u ntp0.zen.co.uk
elif /bin/ping -c 1 ntp2b.mcc.ac.uk ; then
export NETTIME=/usr/sbin/ntpdate -u ntp2b.mcc.ac.uk
elif /bin/ping -c 1 ntp2c.mcc.ac.uk ; then
export NETTIME=/usr/sbin/ntpdate -u ntp2c.mcc.ac.uk
else
export NETTIME=
fi
if [ $NETTIME != ] ; then
sleep 1
sleep 1
$NETTIME
/sbin/clock -u -w
date
fi
which also illustrates how to allow for the possibility that
the default server might not be responding at the time, so
it has 2 fallback servers.
--
Lep pozdrav / With regards,
Gregor Gorjanc
--
University of Ljubljana PhD student
Biotechnical Faculty
Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan
Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si
SI-1230 Domzale tel: +386 (0)1 72 17 861
Slovenia, Europefax: +386 (0)1 72 17 888
--
One must learn by doing the thing; for though you think you know it,
you have no certainty until you try. Sophocles ~ 450 B.C.
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[R] test the tcltk package
Dear all, Could anybody using (K)ubuntu (or Linux in general) confirm if this is a general problem or it's just my box? The problem relates to the Options window in the Rcmdr package (which it looks fine in John Fox's Quantian). The last option (Default font) is stubborn and won't be set; it behaves strangely (e.g. I type 3 and it appears 2). This options should have a button to drag left and right, but in my configuration this is missing; at this link you can see how it looks like: http://www.roda.ro/Options.png What should I do to test if the R tcltk package works fine? I have R 2.3.1, tcl and tk version 8.4 (dev packages installed as well). Thanks in advance, Adrian -- Adrian Dusa Romanian Social Data Archive 1, Schitu Magureanu Bd 050025 Bucharest sector 5 Romania Tel./Fax: +40 21 3126618 \ +40 21 3120210 / int.101 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] test regression against given slope for reduced major axis regression (RMA)
Patrick Drechsler wrote on 11 Jul 2006 02:10:21 MET: [...] I am now confronted with the problem that I have data which requires a modelII regression (also called reduced major axes regression (RMA) or geometric mean regression). For this I use the function modelII (see below). What would be a good way of adapting test_regression_against_slope for use with RMA regression? The question I am trying to answer is: Does the slope acquired from experimental data differ significantly from theoretical predictions? JFTR: David Warton's smatr package solves the problem. DFB: Note that there are conflicting solutions to this question. The confidence limits based on Pitman, EJG (1939) A note on normal correlation. Biometrika 31: 9-12, as implemented in the smatr package, are liberal (i.e. narrow), compared to the alternative limits found in Tan, CY and B. Iglewicz (1999) Measurement-methods comparisons and linear statistical relationship, Technometrics 41: 192-201. Note also that equation error, if present, invalidates any confidence limits calculations. That is, in using the RMA, you are looking for the perfect underlying relationship between two variables. If the underlying relationship is not perfect, because missing variables are also important, the RMA is the wrong approach. -- View this message in context: http://www.nabble.com/test-regression-against-given-slope-for-reduced-major-axis-regression-%28RMA%29-tf1921881.html#a5494654 Sent from the R help forum at Nabble.com. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] test
just a test -- Lep pozdrav / With regards, Gregor Gorjanc -- University of Ljubljana PhD student Biotechnical Faculty Zootechnical Department URI: http://www.bfro.uni-lj.si/MR/ggorjan Groblje 3 mail: gregor.gorjanc at bfro.uni-lj.si SI-1230 Domzale tel: +386 (0)1 72 17 861 Slovenia, Europefax: +386 (0)1 72 17 888 -- One must learn by doing the thing; for though you think you know it, you have no certainty until you try. Sophocles ~ 450 B.C. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test for equality of coefficients in multivariate multipleregression
Dear Berwin, Simply stacking the problems and treating the resulting observations as independent will give you the correct coefficients, but incorrect coefficient variances and artificially zero covariances. The approach that I suggested originally -- testing a linear hypothesis using the coefficient estimates and covariances from the multivariate linear model -- seems simple enough. For example, to test that all three coefficients are the same across the two equations, b - as.vector(coef(x.mlm)) V - vcov(x.mlm) L - c(1, 0, 0,-1, 0, 0, 0, 1, 0, 0,-1, 0, 0, 0, 1, 0, 0,-1) L - matrix(L, nrow=3, byrow=TRUE) t(L %*% b) %*% (L %*% V %*% t(L)) %*% (L %*% b) The test statistic is chi-square with 3 df under the null hypothesis. (Note: not checked carefully.) (BTW, it's a bit unclear to me how much of this exchange was on r-help, but I'm copying to r-help since at least one of Ulrich's messages referring to alternative approaches appeared there. I hope that's OK.) Regards, John John Fox Department of Sociology McMaster University Hamilton, Ontario Canada L8S 4M4 905-525-9140x23604 http://socserv.mcmaster.ca/jfox -Original Message- From: Berwin A Turlach [mailto:[EMAIL PROTECTED] On Behalf Of Berwin A Turlach Sent: Tuesday, July 18, 2006 9:28 PM To: Andrew Robinson Cc: Ulrich Keller; John Fox Subject: Re: [R] Test for equality of coefficients in multivariate multipleregression G'day all, AR == Andrew Robinson [EMAIL PROTECTED] writes: AR I suggest that you try to rewrite the model system into a AR single mixed-effects model, [...] Why a mixed-effect model, wouldn't a fixed effect be o.k. too? Something like: DF-data.frame(x1=rep(c(0,1),each=50),x2=rep(c(0,1),50)) tmp-rnorm(100) DF$y1-tmp+DF$x1*.5+DF$x2*.3+rnorm(100,0,.5) DF$y2-tmp+DF$x1*.5+DF$x2*.7+rnorm(100,0,.5) x.mlm-lm(cbind(y1,y2)~x1+x2,data=DF) coef(x.mlm) y1 y2 (Intercept) -0.08885266 -0.05749196 x1 0.33749086 0.60395258 x2 0.72017894 1.11932077 DF2 - with(DF, data.frame(y=c(y1,y2))) DF2$x11 - with(DF, c(x1, rep(0,100))) DF2$x21 - with(DF, c(x2, rep(0,100))) DF2$x12 - with(DF, c(rep(0,100), x1)) DF2$x22 - with(DF, c(rep(0,100), x2)) DF2$x1 - with(DF, c(x1, x1)) DF2$wh - rep(c(0,1), each=100) fm1 - lm(y~wh + x11 + x21 + x12 + x22, DF2) fm1 Call: lm(formula = y ~ wh + x11 + x21 + x12 + x22, data = DF2) Coefficients: (Intercept) wh x11 x21 x12 x22 -0.08885 0.03136 0.33749 0.72018 0.60395 1.11932 fm2 - lm(y~wh + x1 + x21 + x22, DF2) anova(fm2,fm1) Analysis of Variance Table Model 1: y ~ wh + x1 + x21 + x22 Model 2: y ~ wh + x11 + x21 + x12 + x22 Res.Df RSS Df Sum of Sq F Pr(F) 1195 246.919 2194 246.031 1 0.888 0.6998 0.4039 Cheers, Berwin == Full address Berwin A Turlach Tel.: +61 (8) 6488 3338 (secr) School of Mathematics and Statistics+61 (8) 6488 3383 (self) The University of Western Australia FAX : +61 (8) 6488 1028 35 Stirling Highway Crawley WA 6009e-mail: [EMAIL PROTECTED] Australiahttp://www.maths.uwa.edu.au/~berwin __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test for equality of coefficients in multivariate multipleregression
G'day John, JF == John Fox [EMAIL PROTECTED] writes: JF Simply stacking the problems and treating the resulting JF observations as independent will give you the correct JF coefficients, but incorrect coefficient variances Yes, after Andrew's (off-list) answer I realised this too. If I am not mistaken, all variances/covariances should be off by a factor of 1/2 or something like that. JF and artificially zero covariances. Well, I must admit that I misread Ulrich's code for most of the day. I hadn't realised that the variable `tmp' introduces a correlation between `y1' and `y2' in his code: DF-data.frame(x1=rep(c(0,1),each=50),x2=rep(c(0,1),50)) tmp-rnorm(100) DF$y1-tmp+DF$x1*.5+DF$x2*.3+rnorm(100,0,.5) DF$y2-tmp+DF$x1*.5+DF$x2*.7+rnorm(100,0,.5) for some reason, my brain kept parsing this as generate *one* random intercept for y1 and *one* random intercept for y2, not that each individual observation has a random intercept. Under the model that my brain kept parsing, one would have zero covariances. :) Now I understand why Andrew suggested the use of mixed models and would go down that way too. But I believe your approach is valid too. JF (BTW, it's a bit unclear to me how much of this exchange was JF on r-help, Easy, all those that have r-help either in the TO: or CC: field. Those were Ulrich's original message and the answer by you and Andrew, I kept all my mails so far off-list. JF but I'm copying to r-help since at least one of Ulrich's JF messages referring to alternative approaches appeared there. Yes, I noticed that and answered off-list. In that message, if I read it correctly, he had confused Andrew and me. JF I hope that's OK.) Sure, why not? :) Cheers, Berwin __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test for equality of coefficients in multivariate multiple regression
Hello and thank you for your answers, Andrew and Berwin. If I'm not mistaken, the mixed-model version of Berwin's approach would be: #My stuff: DF-data.frame(x1=rep(c(0,1),each=50),x2=rep(c(0,1),50)) tmp-rnorm(100) DF$y1-tmp+DF$x1*.5+DF$x2*.3+rnorm(100,0,.5) DF$y2-tmp+DF$x1*.5+DF$x2*.7+rnorm(100,0,.5) x.mlm-lm(cbind(y1,y2)~x1+x2,data=DF) #1st part of Andrew's suggestion: DF2 - with(DF, data.frame(y=c(y1,y2))) DF2$x11 - with(DF, c(x1, rep(0,100))) DF2$x21 - with(DF, c(x2, rep(0,100))) DF2$x12 - with(DF, c(rep(0,100), x1)) DF2$x22 - with(DF, c(rep(0,100), x2)) DF2$x1 - with(DF, c(x1, x1)) DF2$wh - rep(c(0,1), each=100) #Mixed version of models: DF2$unit - rep(c(1:100), 2) library(nlme) mm1 - lme(y~wh + x11 + x21 + x12 + x22, random= ~1 | unit, DF2, method=ML) fixef(mm1) (Intercept) wh x11 x21 x12 x22 0.07800993 0.15234579 0.52936947 0.13853332 0.37285132 0.46048418 coef(x.mlm) y1y2 (Intercept) 0.07800993 0.2303557 x1 0.52936947 0.3728513 x2 0.13853332 0.4604842 mm2 - update(mm1, y~wh + x1 + x12 + x22) anova(mm1, mm2) Model df AIC BIClogLik Test L.Ratio p-value mm1 1 8 523.6284 550.0149 -253.8142 mm2 2 7 522.0173 545.1055 -254.0086 1 vs 2 0.388908 0.5329 This seems to be correct. What anova() tells me is that the effect of x1 is the same for y1 and y2. What I don't understand then is why the coefficients for x12 and x22 differ so much between mm1 and mm2: fixef(mm2) (Intercept) wh x1 x12 x22 0.1472766 0.1384474 0.5293695 -0.1565182 0.3497476 fixef(mm1) (Intercept) wh x11 x21 x12 x22 0.07800993 0.15234579 0.52936947 0.13853332 0.37285132 0.46048418 Sorry for being a bit slow here, I'm (obviously) not a statistician. Thanks again, Uli Andrew Robinson wrote: G'day Berwin, my major reason for preferring the mixed-effect approach is that, as you can see below, the residual df for your models are 195 and 194, respectively. The 100 units are each contributing two degrees of freedom to your inference, and presumably to your estimate of the variance. I feel a little sqeueamish about that, because it's not clear to me that they can be assumed to be independent. I worry about the effects on the size of the test. With a mixed-effects model each unit could be a cluster of two observations, and I would guess the size would be closer to nominal, if not nominal. Cheers Andrew __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] Test for equality of coefficients in multivariate multiple regression
Hello,
suppose I have a multivariate multiple regression model such as the
following:
DF-data.frame(x1=rep(c(0,1),each=50),x2=rep(c(0,1),50))
tmp-rnorm(100)
DF$y1-tmp+DF$x1*.5+DF$x2*.3+rnorm(100,0,.5)
DF$y2-tmp+DF$x1*.5+DF$x2*.7+rnorm(100,0,.5)
x.mlm-lm(cbind(y1,y2)~x1+x2,data=DF)
coef(x.mlm)
y1y2
(Intercept) 0.07800993 0.2303557
x1 0.52936947 0.3728513
x2 0.13853332 0.4604842
How can I test whether x1 and x2 respectively have the same effect on y1
and y2? In other words, how can I test if coef(x.mlm)[2,1] is
statistically equal to coef(x.mlm)[2,2] and coef(x.mlm)[3,1] to
coef(x.mlm)[3,2]? I looked at linear.hypothesis {car} and glh.test
{gmodels}, but these do not seem the apply to multivariate models.
Thank you in advance,
Uli Keller
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test for equality of coefficients in multivariate multiple regression
Dear Ulrich,
I'll look into generalizing linear.hypothesis() so that it handles
multivariate linear models.
Meanwhile, vcov(x.mlm) will give you the covariance matrix of the
coefficients, so you could construct your own test by ravelling
coef(x.mlm) into a vector.
I hope that this helps,
John
On Tue, 18 Jul 2006 20:15:12 +0200
Ulrich Keller [EMAIL PROTECTED] wrote:
Hello,
suppose I have a multivariate multiple regression model such as the
following:
DF-data.frame(x1=rep(c(0,1),each=50),x2=rep(c(0,1),50))
tmp-rnorm(100)
DF$y1-tmp+DF$x1*.5+DF$x2*.3+rnorm(100,0,.5)
DF$y2-tmp+DF$x1*.5+DF$x2*.7+rnorm(100,0,.5)
x.mlm-lm(cbind(y1,y2)~x1+x2,data=DF)
coef(x.mlm)
y1y2
(Intercept) 0.07800993 0.2303557
x1 0.52936947 0.3728513
x2 0.13853332 0.4604842
How can I test whether x1 and x2 respectively have the same effect on
y1
and y2? In other words, how can I test if coef(x.mlm)[2,1] is
statistically equal to coef(x.mlm)[2,2] and coef(x.mlm)[3,1] to
coef(x.mlm)[3,2]? I looked at linear.hypothesis {car} and glh.test
{gmodels}, but these do not seem the apply to multivariate models.
Thank you in advance,
Uli Keller
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
John Fox
Department of Sociology
McMaster University
Hamilton, Ontario, Canada
http://socserv.mcmaster.ca/jfox/
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] Test for equality of coefficients in multivariate multiple regression
Hi Uli,
I suggest that you try to rewrite the model system into a single
mixed-effects model, which would allow direct parameterization of the
tests that you're interested in. A useful article, which may be
overkill for your needs, is:
Hall, D.B. and Clutter, M. (2004). Multivariate multilevel nonlinear
mixed effects models for timber yield predictions, Biometrics, 60:
16-24.
See the publications link on Daniel Hall's website:
http://www.stat.uga.edu/~dhall/
Cheers
Andrew
On Tue, Jul 18, 2006 at 08:15:12PM +0200, Ulrich Keller wrote:
Hello,
suppose I have a multivariate multiple regression model such as the
following:
DF-data.frame(x1=rep(c(0,1),each=50),x2=rep(c(0,1),50))
tmp-rnorm(100)
DF$y1-tmp+DF$x1*.5+DF$x2*.3+rnorm(100,0,.5)
DF$y2-tmp+DF$x1*.5+DF$x2*.7+rnorm(100,0,.5)
x.mlm-lm(cbind(y1,y2)~x1+x2,data=DF)
coef(x.mlm)
y1y2
(Intercept) 0.07800993 0.2303557
x1 0.52936947 0.3728513
x2 0.13853332 0.4604842
How can I test whether x1 and x2 respectively have the same effect on y1
and y2? In other words, how can I test if coef(x.mlm)[2,1] is
statistically equal to coef(x.mlm)[2,2] and coef(x.mlm)[3,1] to
coef(x.mlm)[3,2]? I looked at linear.hypothesis {car} and glh.test
{gmodels}, but these do not seem the apply to multivariate models.
Thank you in advance,
Uli Keller
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
--
Andrew Robinson
Department of Mathematics and StatisticsTel: +61-3-8344-9763
University of Melbourne, VIC 3010 Australia Fax: +61-3-8344-4599
Email: [EMAIL PROTECTED] http://www.ms.unimelb.edu.au
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Re: [R] test regression against given slope for reduced major axis regression (RMA)
Patrick Drechsler wrote on 11 Jul 2006 02:10:21 MET: [...] I am now confronted with the problem that I have data which requires a modelII regression (also called reduced major axes regression (RMA) or geometric mean regression). For this I use the function modelII (see below). What would be a good way of adapting test_regression_against_slope for use with RMA regression? The question I am trying to answer is: Does the slope acquired from experimental data differ significantly from theoretical predictions? JFTR: David Warton's smatr package solves the problem. -- You know the world is going crazy when the best rapper is a white guy, the best golfer is a black guy, the Swiss hold the America's Cup, France is accusing the US of arrogance, and Germany doesn't want to go to war. -- Charles Barkley __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] test regression against given slope for reduced major axis regression (RMA)
Hi,
for testing if the slope of experimental data differs from a
given slope I'm using the function
test_regression_against_slope (see below).
I am now confronted with the problem that I have data which
requires a modelII regression (also called reduced major axes
regression (RMA) or geometric mean regression). For this I use
the function modelII (see below).
What would be a good way of adapting
test_regression_against_slope for use with RMA regression?
The question I am trying to answer is: Does the slope acquired
from experimental data differ significantly from theoretical
predictions?
Any feedback on this is highly appreciated. And if you need more
info do not hesitate to ask.
Kind Regards
Patrick
*test_regression_against_slope*
--8schnipp-8---
test_regression_against_slope - function(x,y,slope_2)
{
### TEST_REGRESSION_AGAINST_SLOPE tests a measured regression against a
### given regression.
###
### INPUT:
###
### x and y: raw data
### slope_2: the slope you would like to test against (ie 1/3)
###
### OUTPUT:
###
### pvalue: the P-value...
### upperlimit95 and lowerlimit95 give the 95 percent confidence
### intervall (two-tailed).
###
### see Sokal and Rohlf, p. 465/471
n - length(x)
mydf - n-2
## least square fit:
x2 - (x-mean(x))^2
y2 - (y-mean(y))^2
## regression (pedestrian solution):
xy - (x-mean(x))*(y-mean(y))
slope1 - sum(xy)/sum(x2)
intercept_a - mean(y) - slope1 * mean(x)
## model data y_hat:
y_hat - intercept_a + slope1 * x
## least squares of model data:
y_hat2 - (y - y_hat)^2
s2yx - sum(y_hat2) / (n-2)
sb - sqrt(s2yx/sum(x2))
ts - (slope1 - slope_2) / sb
pvalue - 2*(pt(abs(ts), df, lower.tail=FALSE))
## 0.95 for one-tailed 0.975 for two-tailed t-distribution with
## alpha-5%:
tval - qt(.975, df=mydf)
ts2 - tval*sb
lowerlimit95 - slope1 - ts2
upperlimit95 - slope1 + ts2
list(pvalue = pvalue,
lowerlimit95 = lowerlimit95,
upperlimit95 = upperlimit95)
}
--8schnapp-8---
*modelII*
--8schnipp-8---
modelII - function(XjArray,YjArray){
###
###
### Purpose:
###
### Calculates MODEL II Regression paramaters. Also called reduced
### major axis regression (Prentice 1987) or geometric mean
### regression (Webb et al. 1981).
###
### Input:
###
### Two one dimensional arrays XjArray and YjArray containing the X
### and Y vectors.
###
### XjArray = [0 0.9 1.8 2.6 3.3 4.4 5.2 6.1 6.5 7.4]
### YjArray = [5.9 5.4 4.4 4.6 3.5 3.7 2.8 2.8 2.4 1.5]
###
### Output:
###
### A list with the following:
###
### sumXjYj As Double
### sumXj As Double, sumYj As Double
### sumXjSquared As Double, sumYjSquared As Double
### n As Long
### varXj, varYj
### output(7)
###
###
sumXjYj - 0
sumXj - 0
sumYj - 0
n - 0
n - length(XjArray)
sumXjSquared - 0
sumYjSquared - 0
covariancexy - 0
for(i in 1:n){
sumXjYj - sumXjYj + XjArray[i] * YjArray[i]
sumXj - sumXj + XjArray[i]
sumYj - sumYj + YjArray[i]
sumXjSquared - sumXjSquared + XjArray[i]^2
sumYjSquared - sumYjSquared + YjArray[i]^2
}
## Mean of X and Y vectors
meanyj - sumYj / n
meanxj - sumXj / n
## Create covariance
for(i in 1:n){
covariancexy - covariancexy + ((XjArray[i] - meanxj) * (YjArray[i] -
meanyj))
}
covariancexy - covariancexy / n
## get variance of X and Y (SD)
varXj - (n * sumXjSquared-sumXj^2)/(n*(n - 1))
varYj - (n * sumYjSquared-sumYj^2)/(n*(n - 1))
sdxij - (sumXjSquared)-(sumXj^2/n)
sdxik - (sumYjSquared)-(sumYj^2/n)
## make beta'sgn function to return sign with magnitude of 1
betacoeff - sign(covariancexy) * ((varYj^0.5) / (varXj^0.5))
## 'make intercept
Intercept - meanyj - meanxj * betacoeff
## Make R the pearson produce moment correlation coefficient
if (varYj==0 | varXj==0){
corrCoeff - 0
}else{
corrCoeff - (sumXjYj - ((sumXj * sumYj) / n)) / ((sdxij * sdxik)^0.5)
}
## Make sample variances of betacoefficient and intercept
variancebeta - (varYj / varXj) * ((1 - (corrCoeff ^ 2)) / n)
varianceintercept - (varYj / n) * (1 - corrCoeff) * (2 + ((meanxj ^ 2) * ((1
+ corrCoeff) / varXj)))
sdbeta - variancebeta^0.5
sdintercept - varianceintercept^0.5
list(betacoeff=betacoeff, # - Steigung
Intercept=Intercept,
sdbeta=sdbeta, # standard deviation
sdintercept=sdintercept,
meanxj=meanxj,
meanyj=meanyj,
corrCoeff=corrCoeff) # - pearson correlation koeffizient ;
}
--8schnapp-8---
--
Snoopy (on being house-trained with a rolled-up newspaper):
It does tend however to give one a rather distorted view of the press!
__
[R] Test if an image or random field is stationary
Hi List, I want to know how to test whether an image or generally a random field, is stationary. I am sure there are a lot of intuitive ways to do this but right now I need a formal test which can gives the p-value. And any of those tests already implemented in R? Thanks a lot! Yuefeng [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Test of Significance for overall-accuracy
Hello, I have two classifications. How can I compare the overall-accuracy of these classifications to each other? Is there a possibility within R to test if the achieved overall-accuracy for the classifications are differing significantly? Additionaly, are the McNemar-Test and Broker-Test implented in a package of R? Thank in advance for your help, Markus . Markus Schwarz Wissenschaftlicher Mitarbeiter Eidg. Forschungsanstalt WSL Zürcherstrasse 111 CH-8903 Birmensdorf Telefon +41-44-739 22 87 Fax +41-44-739 22 15 [EMAIL PROTECTED] http://www.wsl.ch/staff/markus.schwarz/ . __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] test for exponential,lognormal and gammadistribution
Hi Nadja It depends on your purpose. Often people are using tests to show that a sample follows a distribution (normal, exponential, lognormal, ...). If a test rejects the null hypothesis that the sample comes from the specified distribution, you are on the safe side, since you are controlling the significance level (e.g. 5%) and therefore know the alpha error. But if a test do not reject the null hypothesis, generally you have NOT shown that the sample has the specified distribution. This is related to the power of your test (to detect differences). If the power of a test is lousy, the conclusion that your sample has the distribution based on the nonsignificant test result is misleading or even wrong. As you mentioned below the kolmogorov-smirnov test does not adapt for the fact that the parameters of the distribution you test against are estimated from the data sample. It assumes that the parameters are know. But in practice that's not the case in general. Since the parameter are estimated from the data, but the test do not have this information, but assumes that these parameters are a fixed known quantity, the test is to conservative and has a small power to detect differences. Therefore it is quite dangerous to conclude that a sample has a specified distribution, based on the kolmogorov-smirnov test. An alternative way might be using graphical tools, e.g. quantile plots (see ?qqplot and ?qqnorm). Obviously you have the same difficulty by interpreting the plots, since nobody can tell you for sure if a deviation from the straight line is significant or just by chance. But if you conclude that a sample has a distribution by looking at the plot you will be aware of this subjectivity that can not be avoided. The test result will often give you the wrong impression of objectivity. The best example to see this is if you have a very small sample. In general any test has a small power if your sample is small and it is most probable that the test is nonsignificant. If we look at the quantile plot (with a small sample) we often can not judge if it is a straight line or not (since the sample is to small) and in this case it is the correct conclusion that we can not say anything about the distribution. I hope this will be helpful. Regards, Christoph Buser -- Christoph Buser [EMAIL PROTECTED] Seminar fuer Statistik, LEO C13 ETH (Federal Inst. Technology) 8092 Zurich SWITZERLAND phone: x-41-44-632-4673 fax: 632-1228 http://stat.ethz.ch/~buser/ -- [EMAIL PROTECTED] writes: hello! i don't want to test my sample data for normality, but exponential- lognormal- or gammadistribution. as i've learnt the anderson-darling-test in R is only for normality and i am not supposed to use the kolmogorov-smirnov test of R for parameter estimates from sample data, is that true? can you help me, how to do this anyway! thank you very much! nadja __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html !DSPAM:43219660106581956711619! __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] test for exponential,lognormal and gammadistribution
I just got 73 hits for 'RSiteSearch(test for exponential distribution)'. In skimming quickly the first 20, the one that seemed most relevant to me was a reply I wrote to a similar question two months ago (http://finzi.psych.upenn.edu/R/Rhelp02a/archive/37055.html). Of course, some of the remaining 72 may help you more. spencer graves [EMAIL PROTECTED] wrote: hello! i don't want to test my sample data for normality, but exponential- lognormal- or gammadistribution. as i've learnt the anderson-darling-test in R is only for normality and i am not supposed to use the kolmogorov-smirnov test of R for parameter estimates from sample data, is that true? can you help me, how to do this anyway! thank you very much! nadja __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Spencer Graves, PhD Senior Development Engineer PDF Solutions, Inc. 333 West San Carlos Street Suite 700 San Jose, CA 95110, USA [EMAIL PROTECTED] www.pdf.com http://www.pdf.com Tel: 408-938-4420 Fax: 408-280-7915 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] test for exponential,lognormal and gammadistribution
hello! i don't want to test my sample data for normality, but exponential- lognormal- or gammadistribution. as i've learnt the anderson-darling-test in R is only for normality and i am not supposed to use the kolmogorov-smirnov test of R for parameter estimates from sample data, is that true? can you help me, how to do this anyway! thank you very much! nadja __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] test for difference in the order of self generated sequence
Hello, I have an experiment in which we ask subjects to generate a list of thoughts after being exposed to a stimuli. The thoughts were then coded into two categories (e.g. A B). The objective is to show that the order in which thoughts are generated is affected by the experimental conditions. Can somebody tell me what functions in R can help me do the test or point me to the appropriate reference? Thanks! Example data: Subject 1 in condition 1 generated 5 thoughts. The first, second, and 5th thought is in category A, thought 3 and 4 are in category B. Subject 2 only generated 4 thoughts. ---Condition-A-B n1 11, 2, 5 3, 4 n2 23, 4 1, 2 --- Kaiya Liu __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] test for difference in the order of self generated sequence
Have you considered glmmPQL in library(MASS)? spencer graves Kaiya Liu wrote: Hello, I have an experiment in which we ask subjects to generate a list of thoughts after being exposed to a stimuli. The thoughts were then coded into two categories (e.g. A B). The objective is to show that the order in which thoughts are generated is affected by the experimental conditions. Can somebody tell me what functions in R can help me do the test or point me to the appropriate reference? Thanks! Example data: Subject 1 in condition 1 generated 5 thoughts. The first, second, and 5th thought is in category A, thought 3 and 4 are in category B. Subject 2 only generated 4 thoughts. ---Condition-A-B n1 11, 2, 5 3, 4 n2 23, 4 1, 2 --- Kaiya Liu __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Spencer Graves, PhD Senior Development Engineer PDF Solutions, Inc. 333 West San Carlos Street Suite 700 San Jose, CA 95110, USA [EMAIL PROTECTED] www.pdf.com http://www.pdf.com Tel: 408-938-4420 Fax: 408-280-7915 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] test of hazard ratios
Can you fit a single model that includes the two classifiers from which the model with one classifier could be obtained as a special case? If yes (and if neither parameter was at a boundary crudely similar to a zero variance component in lme), then you could use the theory that 2*ln(likelihood ratio) is approximately chi-square. For many capabilities in R, this is implemented as a function anova(fit2, fit1) where fit2 and fit1 are the results of fitting the general and the specialized models. If you don't see how to make this work, PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html;. It might help you find a better answer. Failing that, if you submit another post after following the process described therein, it should increase the chances that you will get a useful reply. spencer graves Steve Adams wrote: Hi, Is there a R test available that tests whether 2 hazard ratios obtained from Cox regressions on the same patient sample by 2 different classifiers are significantly different? Thanks Steve __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Spencer Graves, PhD Senior Development Engineer PDF Solutions, Inc. 333 West San Carlos Street Suite 700 San Jose, CA 95110, USA [EMAIL PROTECTED] www.pdf.com http://www.pdf.com Tel: 408-938-4420 Fax: 408-280-7915 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] test of hazard ratios
Hi, Is there a R test available that tests whether 2 hazard ratios obtained from Cox regressions on the same patient sample by 2 different classifiers are significantly different? Thanks Steve __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] test for equality of two data sets with multidimensional variables
Hello there, I have two data sets with 14 variables each, and wish to do the test for equality of their covariance matrices and mean vectors. Normally these tests should be done by chi square test (box provided) and Hotelling's T square test respectively. Which R functions could do this kind of test? I just find some functions could do for one dimension, but no for multidimension. Some one suggests bartlett.test, but it seems just works for one dimension. Do you know which ones could do that, or I have to do R programming by myself? Thank you, Shengzhe __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] test for equality of two data sets with multidimensional variables
Hi searching in CRAN homepage for hotelling gave me this response Thanks everyone for your help on this question. I solved the problem by writing a procedure to calculate Hotelling's T^2 for a one-sample, multivariate t-test. Here's how it looks, perhaps it will be useful to others. data - cbind(rnorm(50, 0.1, .01), rnorm(50,.1,.01), rnorm(50,.1,.01)) k - ncol(data) n - nrow(data) xbar - apply(data, 2, mean) mubar - rep(0,k) #hypothesized means are zero dbar - xbar - mubar v - var(data) t2 - n*dbar%*%solve(v)%*%dbar F - (n-k)*t2/((n-1)*k) P - 1-pf(F,k,n-k) A previous post by Peter B. Mandeville was very helpful, as well as the Johnson/Wichern book on multivariate stats. -S. Schultz and this cran.r-project.org/doc/packages/agce.pdf - Podobné stránky CRAN - Package SharedHT2 SharedHT2: Shared Hotelling T2 test for small sample microarray experiments ... Derives a Hotelling T2 statistic having an F-distribution using an empirical ... Maybe this is what you want. HTH Petr On 21 Jun 2005 at 13:00, wu sz wrote: Hello there, I have two data sets with 14 variables each, and wish to do the test for equality of their covariance matrices and mean vectors. Normally these tests should be done by chi square test (box provided) and Hotelling's T square test respectively. Which R functions could do this kind of test? I just find some functions could do for one dimension, but no for multidimension. Some one suggests bartlett.test, but it seems just works for one dimension. Do you know which ones could do that, or I have to do R programming by myself? Thank you, Shengzhe __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Petr Pikal [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] test for equality of two data sets with multidimensional variables
Hi, I just find another package shapes for solving the Hotelling T^2 test, but what about the R function for the test of covariance equality(chi ^ 2 box's test) between two multidimensional data sets? Shengzhe __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] test for equality of two data sets with multidimensional variables
wu sz [EMAIL PROTECTED] writes: Hi, I just find another package shapes for solving the Hotelling T^2 test, but what about the R function for the test of covariance equality(chi ^ 2 box's test) between two multidimensional data sets? Shengzhe Rather than looking for a package, it might be more expedient to grab TW Anderson's book and code the couple of lines it takes to generate the test statistic and its asymptotic p-value. -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] test for equality of two data sets withmultidimensional variables
Dear Petr and Shengzhe, Also see ?manova. John John Fox Department of Sociology McMaster University Hamilton, Ontario Canada L8S 4M4 905-525-9140x23604 http://socserv.mcmaster.ca/jfox -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Petr Pikal Sent: Tuesday, June 21, 2005 7:30 AM To: wu sz; r-help@stat.math.ethz.ch Subject: Re: [R] test for equality of two data sets withmultidimensional variables Hi searching in CRAN homepage for hotelling gave me this response Thanks everyone for your help on this question. I solved the problem by writing a procedure to calculate Hotelling's T^2 for a one-sample, multivariate t-test. Here's how it looks, perhaps it will be useful to others. data - cbind(rnorm(50, 0.1, .01), rnorm(50,.1,.01), rnorm(50,.1,.01)) k - ncol(data) n - nrow(data) xbar - apply(data, 2, mean) mubar - rep(0,k) #hypothesized means are zero dbar - xbar - mubar v - var(data) t2 - n*dbar%*%solve(v)%*%dbar F - (n-k)*t2/((n-1)*k) P - 1-pf(F,k,n-k) A previous post by Peter B. Mandeville was very helpful, as well as the Johnson/Wichern book on multivariate stats. -S. Schultz and this cran.r-project.org/doc/packages/agce.pdf - Podobné stránky CRAN - Package SharedHT2 SharedHT2: Shared Hotelling T2 test for small sample microarray experiments ... Derives a Hotelling T2 statistic having an F-distribution using an empirical ... Maybe this is what you want. HTH Petr On 21 Jun 2005 at 13:00, wu sz wrote: Hello there, I have two data sets with 14 variables each, and wish to do the test for equality of their covariance matrices and mean vectors. Normally these tests should be done by chi square test (box provided) and Hotelling's T square test respectively. Which R functions could do this kind of test? I just find some functions could do for one dimension, but no for multidimension. Some one suggests bartlett.test, but it seems just works for one dimension. Do you know which ones could do that, or I have to do R programming by myself? Thank you, Shengzhe __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Petr Pikal [EMAIL PROTECTED] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] test of thread behaviour - no need to read this
test of thread behaviour this is the initial post __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] test of thread behaviour - no need to read this
this is the first reply to the initial thread subject line and header unchanged __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] test of thread behaviour - no need to read this -- version2
this is the second reply to the initial thread subject line has been changed __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] test of thread behaviour - no need to read this -- version 3
this is the fourth reply to the initial post subject line edited ( Re was removed ) __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] test of thread behaviour - no need to read this
this is the seventh reply to the initial post subject line is unaltered I added a ficticious (i think) email to the To: field __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Test on mu with multivariate normal distribution
Telse Henschel [EMAIL PROTECTED] writes: Dear WizaRds, I am sorry to bother you with a newbie question, but although I tried to solve my problem using the various .pdf files (Introduction, help pages etc.), I have come to a complete stop. Please be so kind as to guide me a little bit along my way of exploring multivariate analysis in R. I want to test wether the means-vector mu1 of X, consisting of the means per column of that matrix , and mu2, i.e. the means per column of Y, are distributed equally under the assumption of a multivariate normal distribution. I thought simtest could be the right function to get the p-value, but I fail to use R correctly. Here is what I tried to do: f1- factor(c(8, 10, 12, 14)) X - matrix(1:16, ncol=4) colnames(X) - c(Obj1,Obj2,Obj3,Obj4) X.frame - data.frame(f1,X) Y - matrix(1:12, ncol=4) colnames(Y) - c(Obj1,Obj2,Obj3,Obj4) Y.frame - data.frame(f1,Y) XY- data.frame(X.frame, Y.frame) # wont work, because of different nr of rows test.stat - lm(XY ~ f1) # ??? Try this: X - matrix(rnorm(16), ncol=4) Y - matrix(rnorm(12), ncol=4) XY - rbind(X,Y) g - factor(rep(1:2,c(4,3))) fit1 - lm(XY ~ g) fit0 - lm(XY ~ 1) anova(fit0,fit1) (using 1:16 and 1:12 for X and Y gives Error in anova.mlmlist(object = fit0, fit1) : residuals have rank 1 4 ) # simtest(test.stat, XY, type=Tukey) creates errors already by just staring at it. I apologize. Thank you so much for your support, Telse __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- O__ Peter Dalgaard Blegdamsvej 3 c/ /'_ --- Dept. of Biostatistics 2200 Cph. N (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Test on mu with multivariate normal distribution
Dear WizaRds, I am sorry to bother you with a newbie question, but although I tried to solve my problem using the various .pdf files (Introduction, help pages etc.), I have come to a complete stop. Please be so kind as to guide me a little bit along my way of exploring multivariate analysis in R. I want to test wether the means-vector mu1 of X, consisting of the means per column of that matrix , and mu2, i.e. the means per column of Y, are distributed equally under the assumption of a multivariate normal distribution. I thought simtest could be the right function to get the p-value, but I fail to use R correctly. Here is what I tried to do: f1 - factor(c(8, 10, 12, 14)) X - matrix(1:16, ncol=4) colnames(X) - c(Obj1,Obj2,Obj3,Obj4) X.frame - data.frame(f1,X) Y - matrix(1:12, ncol=4) colnames(Y) - c(Obj1,Obj2,Obj3,Obj4) Y.frame - data.frame(f1,Y) XY - data.frame(X.frame, Y.frame) # wont work, because of different nr of rows test.stat - lm(XY ~ f1) # ??? # simtest(test.stat, XY, type=Tukey) creates errors already by just staring at it. I apologize. Thank you so much for your support, Telse __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Test for autocorrelation in nlme model
Dear all, I am fitting a nonlinear mixed-effects model from a balanced panel of data using nlme. I would like to know whay would be the best options for formally testing for autocorrelation. Is it possible to carry out a Durbin-Watson test on a nlme object? As far as I've seen, I think the durbin.watson function from the car package just works on lm objects. Thank you very much, Antonio __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Test for autocorrelation in nlme model
Have you reviewed Pinheiro and Bates (2000) Mixed-Effects Models in S and S-PLUS (Springer)? They provide examples with diagnostic plots, confidence intervals and likelihood ratio tests in sec. 5.3, for example. spencer graves Revilla,AJ (pgt) wrote: Dear all, I am fitting a nonlinear mixed-effects model from a balanced panel of data using nlme. I would like to know whay would be the best options for formally testing for autocorrelation. Is it possible to carry out a Durbin-Watson test on a nlme object? As far as I've seen, I think the durbin.watson function from the car package just works on lm objects. Thank you very much, Antonio __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] test ignore
__ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] [TEST] Do not read
test __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] test of significance for nlme coefficients
Hello, I am using the nlme package to fit an exponential decay model to a longitudinal data set: ##Model a variable called RASCH16A as an exponential function ## of an initial value A and a decay rate K fm2 - nlme(model = RASCH16A ~ A* (exp(-K*DAYS)), + ###A has nonzero fixed and random effects, K has nonzero fixed and random effects + fixed = A + K ~ 1, + data = rasch, + random = A + K ~ 1 | SUBJECT, + ##Give a start value for the fixed effects, expressed using the concatenate function + start = c(60, 0.007)) However, 20 of the 161 estimates decay rates are NOT decay rates, but are very small positive growth rates (all less than a standard deviation (.03) from zero). I would like to run a significance test to see how many of these positive growth rates are significantly different from zero. However, the nlme package only offers confidence intervals for the model parameters, not the person-specific coefficients. I would welcome any and all suggestions. Thanks if you can help, Lindsey Root __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] test comparing spatial point patterns?
I'm working with animal relocation data. These are x-y coordinates of radio-collared coyotes over a period of time.Relocations are far enough apart in time to be considered independent. We want to test if the pattern of space-use has changed from one year to the next. Relocations roughly follow a 2D Gaussian distribution, but points are often clustered near one or more 'cores'. Sample size varies from year to year with a minimum of n=34. We have applied multi-response permutation procedures (MRPP), but feel this test is too sensitive to the density of points (vs location). We have considered using a 2D generalization of the Cramer-von Mises test (to be adapted from Syrjala, 1996), but this test is sensitive to placement of the origin and probably requires gridding of the data. Does anybody know of a good method to compare the spatial distribution of point pattern data? Cheers, Alan Swanson [[alternative HTML version deleted]] __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] test multiple objects for being equal length
I could not find any help pages on How to test many objects for being of equal length Something like identical for more than two objects? x-1:6 y-1:10 z-3:5 ## For two objects I can do: identical(length(x),length(y)) ## For more than two I currently can do: length(unique(c(length(x),length(y),length(z==1 but there must be a better way. Thanks, M __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] test multiple objects for being equal length
Hi Manuel, First, encapsulate yoyr objects within a list. That will help you manipulate all them at once and ensures that the final function will work with whatever number of vectors. » ll - list(x,y,z) » sapply(ll,length) [1] 6 10 3 Then you can use your length(unique(...))==1 Another way is to use all: » all(sapply(ll,length)==length(ll[[1]])) [1] FALSE HTH, Eric At 09:09 9/12/2004, Manuel Gutierrez wrote: I could not find any help pages on How to test many objects for being of equal length Something like identical for more than two objects? x-1:6 y-1:10 z-3:5 ## For two objects I can do: identical(length(x),length(y)) ## For more than two I currently can do: length(unique(c(length(x),length(y),length(z==1 but there must be a better way. Thanks, M __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Eric Lecoutre UCL / Institut de Statistique Voie du Roman Pays, 20 1348 Louvain-la-Neuve Belgium tel: (+32)(0)10473050 [EMAIL PROTECTED] http://www.stat.ucl.ac.be/ISpersonnel/lecoutre If the statistics are boring, then you've got the wrong numbers. -Edward Tufte __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] test multiple objects for being equal length
not that its much shorter: length( table( sapply( list(x,y,z), length ) ) ) == 1 Bendix -- Bendix Carstensen Senior Statistician Steno Diabetes Center Niels Steensens Vej 2 DK-2820 Gentofte Denmark tel: +45 44 43 87 38 mob: +45 30 75 87 38 fax: +45 44 43 07 06 [EMAIL PROTECTED] www.biostat.ku.dk/~bxc -- -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Manuel Gutierrez Sent: Thursday, December 09, 2004 9:09 AM To: [EMAIL PROTECTED] Subject: [R] test multiple objects for being equal length I could not find any help pages on How to test many objects for being of equal length Something like identical for more than two objects? x-1:6 y-1:10 z-3:5 ## For two objects I can do: identical(length(x),length(y)) ## For more than two I currently can do: length(unique(c(length(x),length(y),length(z==1 but there must be a better way. Thanks, M __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] test multiple objects for being equal length
Wrapping the suggestions into a function:
sameLength - function(...) {
+ n - sapply(list(...), length)
+ all(n == n[1])
+ }
sameLength(double(1), double(2))
[1] FALSE
sameLength(double(1), double(1), list(x=1))
[1] TRUE
[Note that if you have lots of objects to compare, length(unique(...))==1
will not be as efficient.]
HTH,
Andy
From: Manuel Gutierrez
I could not find any help pages on How to test many
objects for being of equal length
Something like identical for more than two objects?
x-1:6
y-1:10
z-3:5
## For two objects I can do:
identical(length(x),length(y))
## For more than two I currently can do:
length(unique(c(length(x),length(y),length(z==1
but there must be a better way.
Thanks,
M
__
[EMAIL PROTECTED] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
__
[EMAIL PROTECTED] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] test
[[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] test
[[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Test of significance in estimation of correlation coefficients
I estimated spearman's correlation coefficient using cor(). How do I test for significance? Vikas __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Test of significance in estimation of correlation coefficients
On 08/24/04 15:05, Vikas Rawal wrote: I estimated spearman's correlation coefficient using cor(). How do I test for significance? cor.test You could find this many ways, but one is to look at the help document for cor. cor.test is listed is at the bottom. In general, it is a good idea to look at help before posting here. BTW, there is a small bug in the documentation for cor: cor.test is no longer in the ctest package, but in stats. Jon -- Jonathan Baron, Professor of Psychology, University of Pennsylvania Home page: http://www.sas.upenn.edu/~baron R search page: http://finzi.psych.upenn.edu/ __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Test of significance in estimation of correlation coefficients
On Tue, 24 Aug 2004, Jonathan Baron wrote: BTW, there is a small bug in the documentation for cor: cor.test is no longer in the ctest package, but in stats. Your information is not up to date: that page has been changed in R-patched and R-devel. -- Brian D. Ripley, [EMAIL PROTECTED] Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272866 (PA) Oxford OX1 3TG, UKFax: +44 1865 272595 __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] test for difference between non-independent correlations
Hello, I am wondering whether there is a way to test whether two non-independent correlation coefficients are significantly different, in R? I have an experimentally measure variable Y, and two different variables X1, and X2, which are predictions of Y that were predicted using two different computational models. I would like to see whether the correlation of Y and X1, and Y and X2 is significantly different. Since Y appears in both correlations the correlation coefficients will be non-independent. (so you cannot use a Z test like you would for comparing independent correlation coefficients, I think). I have read that there is a test for comparing non-independent correlation coefficients, by EJ Williams (1959) and Steiger. There is also a test called Hotelling's t test. I'm wondering does anyone know whether I can somehow carry out any of these using an R package, or compare non-independent correlations using R by some other method? I will greatly appreciate any help, thankyou, Avril Coghlan (University College Dublin, Ireland) __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] test for difference between non-independent correlations
On 07/27/04 13:35, Avril Coghlan wrote: Hello, I am wondering whether there is a way to test whether two non-independent correlation coefficients are significantly different, in R? About 20 years ago I asked the same question (without the R), found Steiger's paper (below) and wrote this BASIC program to do it. (If you try to type it in each time you're bound to make a mistake. This has been checked repeatedly.) It is trivial to write this as an R script, of course. I and my colleagues have been using this for about 20 years, and it seems to be OK as judged by comparing it to the jackknife. (Remember, when we did this, the bootstrap was a newfangled idea that we didn't quite trust.) My hunch is that something better has come along, and I'd be interested to see how others respond. Hence I haven't bothered to translate this into R. Or maybe everyone who has this problem just uses the bootstrap now. That is certainly easy enough. (I will eventually add this to our Notes on R for psychology..., since it is a very common problem in psychology research.) 10 print looks for difference between r12 and r13, given r23 12 print (input r121 to quit) 20 input r12 ; r12 22 if r121 then end 30 input r13 ; r13 40 input r23 ; r23 50 input N ; n 52 let rd=1-r12*r12-r13*r13-r23*r23+2*r12*r13*r23 54 let rb=(r12+r13)/2 56 let cube=(1-r23)*(1-r23)*(1-r23) 60 let t2=(r12-r13)*sqrt((n-1)*(1+r23)/(2*rd*(n-1)/(n-3)+rb*rb*cube)) 70 print t (;n-3;) = ;t2 80 print Steiger, J.H. (1980). Tests for comparing elements of a 82 print correlation matrix. Psychological Bulletin, 87, 245-251. 100 goto 12 -- Jonathan Baron, Professor of Psychology, University of Pennsylvania Home page: http://www.sas.upenn.edu/~baron R search page: http://finzi.psych.upenn.edu/ __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] test for end of file on connection
Hi, I am looking for a function to test for end-of-file on a connection. Apparently this question was already asked a couple of years ago and then P. Dalgaard suggested to look at help(connections), help(readLines). Unfortunately, I couldn't find such a function on those pages, maybe I am missing something. Did anyone figure this out? Thanks, Vadim [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] test for end of file on connection
With the text of your message copied to the clipboard: con - file(clipboard, r) readLines(con, 1) [1] I am looking for a function to test for end-of-file on a connection. readLines(con, 1) [1] Apparently this question was already asked a couple of years ago and readLines(con, 1) [1] then P. Dalgaard suggested to look at help(connections), readLines(con, 1) [1] help(readLines). Unfortunately, I couldn't find such a function on those readLines(con, 1) [1] pages, maybe I am missing something. readLines(con, 1) character(0) i.e., readLines() returns a zero length result upon reaching end of file. AFAIK the other file reading functions have similar behavior. It's still worth reading in detail the help for readLines(). hope this helps, Tony Plate At Tuesday 12:08 AM 5/11/2004, Vadim Ogranovich wrote: Hi, I am looking for a function to test for end-of-file on a connection. Apparently this question was already asked a couple of years ago and then P. Dalgaard suggested to look at help(connections), help(readLines). Unfortunately, I couldn't find such a function on those pages, maybe I am missing something. Did anyone figure this out? Thanks, Vadim [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
