On Aug 6, 2011, at 03:23 , Walmes Zeviani wrote:
Hi,
I've data from an incomplete fatorial design. One level of a factor doesn't
has the levels of the other. When I use lm(), the summary() return NA for
that non estimable parameters. Ok, I understant it. But I use
contrast::contrast(),
On Fri, 05-Aug-2011 at 02:32PM +0200, peter dalgaard wrote:
|
| On Aug 5, 2011, at 14:10 , Alexander Engelhardt wrote:
|
| Am 05.08.2011 12:24, schrieb Patrick Connolly:
| I don't wish to install R by rpm. I need to know what Fedora rpms I
| need to install to give me the capability to
Hi,
this is my script
anova - aov(data ~ Ts*Te*t + Error(R/Ts*Te*t))
results - summary(anova$Within)
this is results
Df Sum Sq Mean Sq F valuePr(F)
Ts 2 1232.2 616.11 53.606 3.965e-10 ***
Ts:Te 4 4889.5 1222.37 106.356 4.075e-16 ***
Ts:t 4 6472.1
On 08/06/2011 03:46 AM, SavageMaDaMe wrote:
Hi- I am trying to plot a matrix of categorical values across time using
color to represent each individual factor. For example:
1982 1983 1984 1985 1986 1987
119 19 68 68 19 19 68
268 68 19 19 68 68 19
326 26
Hi there
Would you be so kind to send an example, please?
Many thanks
Ed
On Fri, Aug 5, 2011 at 9:43 PM, Duncan Mackay mac...@northnet.com.auwrote:
Hi
The are three places where the size of the graphs can be customised
1 Sweave with options
2 R in the device
3 Latex specifying
On 5 August 2011 at 23:19, Barnet Wagman wrote:
| Has anyone successfully installed rJava on Debian squeeze (or any other
| version of linux)?
|
| I keep getting the
|
| ... (JDK) is missing or not registered in R
|
|
| error message when installing the package.
|
| I've tried running
It looks like you want
results[[1]]$Df[5]# residual df
results[[1]]$`Mean Sq`[5]
Untested without a reproducible example.
Dennis
On Sat, Aug 6, 2011 at 3:00 AM, Roberto rmosce...@unitus.it wrote:
Hi,
this is my script
anova - aov(data ~ Ts*Te*t + Error(R/Ts*Te*t))
results -
On Aug 6, 2011, at 6:00 AM, Roberto wrote:
Hi,
this is my script
anova - aov(data ~ Ts*Te*t + Error(R/Ts*Te*t))
results - summary(anova$Within)
this is results
Df Sum Sq Mean Sq F valuePr(F)
Ts 2 1232.2 616.11 53.606 3.965e-10 ***
Ts:Te 4 4889.5 1222.37
Hi,
I'm a new user of R. I want to install 'copula' packages. I downloaded the
package and from the tab above in R-console i selected install packages
from Zipped file. After this when I 'm writing in a command prompt
library(copula). It gives the error massage as
Error in library(copula) :
Dear List,
I am a beginner of R and have an easy question, which I couldnt find out.
I like to exclude all rows of matrix y from matrix x (like a subset of
x, without y).
The matrix x is of the structure
str(x)
num [1:346, 1:8] 0.055 0.6833 0.9121 0.0819 0.1223 ...
-
If you know the names will always work (i.e., names of y are always names of
x and the rows always have names) this should do it:
x[!(rownames(x) %in% rownames(y)),]
I.e., those rows of x such that their names are NOT included in the row
names of y.
Hope this helps,
Michael Weylandt
On Sat,
Exactly right Peter. Thanks.
There should be some way for me to detect such situations so as to not
result in an impressive P-value. Ideas welcomed!
This is a great example why users should post a toy example on the first
posting, as we can immediately see that this model MUST fit the data, so
Hi,
1) What happened when you installed copula? Did you get any error or
warning messages?
2) Do you have all of the dependencies installed? At least for the
most recent version of copula on CRAN they are: methods, mvtnorm,
scatterplot3d, sn, pspline.
3) What version of R are you using? (you
Hi R users,
I sent a message yesterday about NA in model estimates (
http://r.789695.n4.nabble.com/How-set-lm-to-don-t-return-NA-in-summary-td3722587.html).
If I use aov() instead of lm() I get no NA in model estimates and I use
gmodels::estimable() without problems. Ok!
Now I'm performing a lot
Hi Walmes,
The problem goes back to the singularity caused by the incomplete
design. The summary() and coef() methods for aov class objects hide
it rather than printing NA as with lm(), but look at just:
m1
it says at the end 1 out of 6 effects not estimable. Now when you
use glht() you pass
This is a nice solution. Thanks, Dennis. But I am afraid if the length
of the list x isn't equal to the length of x2, there will be errors
since lapply returns a list of the same length.
x - list(A=c(d, e, f), B=c(d, e), C=c(d,g))
x2 - unique(unlist(x))
w - lapply(x, function(u)
(Dirk: I mistakenly sent this to you directly: I meant to post it to R
help. Sorry)
On 8/6/11 8:16 AM, Dirk Eddelbuettel wrote:
a) Did you try 'sudo apt-get install r-cran-rjava' ? This is after all
packaged in Debian
b) The Debian rjava package has 'openjdk-6-jdk' in its
Dear list,
I'm currently trying to use the rms package to get predicted ordinal
responses from a conditional ratio model. As you will see below, my
model seems to fit well to the data, however, I'm having trouble
getting predicted mean (or fitted) ordinal response values using the
predict
On 8/6/11 9:37 AM, Dirk Eddelbuettel wrote:
| I'm going to be using rJava as part of a package I'm developing, so I
| need to be able to install it in a more generic way than the Debian
Care to define more generic way than Debian ?
^ I mean:
(i) Install Java from Oracle
(ii) run R CMD
Unfortunately the list names of my real data are irregular with mixed
digit and letters at the end. This is good idea though. It inspired me
to give another solution based on that:
x - list(A=c(d, e, f), B=c(d, e), C=c(d,g))
tmp - unlist(x, use.names=F)
a = unlist(lapply(x, length))
tmp2 =
If you have an Internet connection try:
install.packages(copula, dep=TRUE)
Good luck!
Ken
On Aug 6, 2554 BE, at 8:44 AM, poulomi poulomi...@gmail.com wrote:
Hi,
I'm a new user of R. I want to install 'copula' packages. I downloaded the
package and from the tab above in
Dear R-users,
I am comparing differences in variance, skew, and kurtosis between two groups.
For variance the comparison is easy: just
var.test(group1, group2)
I am using agostino.test() for skew, and anscombe.test() for kurtosis. However,
I can't find an equivalent of the F.test or
For an application of ridge regression, I need to get the covariance
matrices of the estimated regression
coefficients in addition to the coefficients for all values of the ridge
contstant, lambda.
I've studied the code in MASS:::lm.ridge, but don't see how to do this
because the code is
The help files do not make this clear, but predict and Mean in rms do not
support computation of predicted means for the continuation ratio model.
The help file for cr.setup shows you how to compute the unconditional
probabilities you would need for doing that.
Frank
apeer wrote:
Dear list,
Hey,
I have a set of income data which I'd like to fit to a gamma
distribution. How can I estimate the two parameters of the gamma
distribution for a vector, e.g.
c(2039L, 2088L, 5966L, 2353L, 1966L, 2312L, 3305L, 2013L, 3376L,
3363L, 3567L, 4798L, 2032L, 1699L, 3001L, 2329L, 3944L, 2568L,
Hi Alex,
Try
require(MASS)
Loading required package: MASS
b - c(2039L, 2088L, 5966L, 2353L, 1966L, 2312L, 3305L, 2013L, 3376L,
+ 3363L, 3567L, 4798L, 2032L, 1699L, 3001L, 2329L, 3944L, 2568L,
+ 1699L, 4545L)
fitdistr(b, 'gamma')
shape rate
6.4528939045 0.0021887943
On Fri, Aug 05, 2011 at 11:47:49PM +0530, Ron Michael wrote:
Hi all, I have happened to work on MS .NET for sometime now, and I found that
this language offers RNG what is called as Donald E. Knuth's subtractive
random number generator algorithm (found here:
On 08/06/2011 09:23 PM, Jorge Ivan Velez wrote:
Hi Alex,
Try
require(MASS)
Loading required package: MASS
b- c(2039L, 2088L, 5966L, 2353L, 1966L, 2312L, 3305L, 2013L, 3376L,
+ 3363L, 3567L, 4798L, 2032L, 1699L, 3001L, 2329L, 3944L, 2568L,
+ 1699L, 4545L)
fitdistr(b, 'gamma')
shape
On 8/6/11 9:37 AM, Dirk Eddelbuettel wrote:
| I'm going to be using rJava as part of a package I'm developing, so I
| need to be able to install it in a more generic way than the Debian
Care to define more generic way than Debian ?
^ I mean:
(i) Install Java from Oracle
(ii) run R CMD
(Dirk: I mistakenly sent this to you directly: I meant to post it to R
help. Sorry)
On 8/6/11 8:16 AM, Dirk Eddelbuettel wrote:
a) Did you try 'sudo apt-get install r-cran-rjava' ? This is after all
packaged in Debian
b) The Debian rjava package has 'openjdk-6-jdk' in its
Thanks, Frank. As a rule, I provide an example in my first post.
However, in this case, the data are confidential, and I was not
allowed to provide you those data. Moreover, I thought that I was not
able to generate data exhibiting the reported problem.
Paul
On Sat, Aug 6, 2011 at 3:27 PM,
Hi all, I'm learning R. I'm used to Matlab and am having the following issue:
I define a column vector and matrix and then multiply them. The result is
not what I expect:
v2 - c(0,1,0)
M - cbind(c(1,4,7),c(2,5,8),c(3,6,9))
M*v2
[,1] [,2] [,3]
[1,]000
[2,]456
[3,]
Hello,
This might seem a trivial and stupid question. I have generated a large file
with results after landmark reflection, and every time I try saving it, it
only saves the part which is displayed in the R consoleHow can I get my
results for all processed specimens?
Thank you in advance
On Aug 6, 2011, at 4:35 PM, smajor wrote:
Hi all, I'm learning R. I'm used to Matlab and am having the
following issue:
I define a column vector and matrix and then multiply them. The
result is
not what I expect:
v2 - c(0,1,0)
M - cbind(c(1,4,7),c(2,5,8),c(3,6,9))
M*v2
[,1] [,2]
Hi,
You are using the regular multiplication operator, when you want the
matrix multiplication operator. Compare:
M * v2
[,1] [,2] [,3]
[1,]000
[2,]456
[3,]000
M %*% v2
[,1]
[1,]2
[2,]5
[3,]8
See ?* and ?%*%
HTH,
Josh
On Sat, Aug
Hi Jana,
How are you trying to save the file you created? Depending on what
sort of R object you have (data frame/matrix or list or ...), try
looking at ?write.table and ?save as some possibilities for keeping
your data. In the future, if you provide us with more details (like
the version of R
You need to learn that R syntax is not the same as Matlab syntax.
If you want to use R, learn R. Read the basic introductory material,
readily available from the R web site.
In particular R treats multiplication --- i.e. * --- as multiplication.
Arrays are multiplied entry by entry when they
By default, R operations on matrices, vectors, etc are elementwise like .* in
Matlab. If you want matrix multiplication use %*%
Hope this helps and good luck with R!
Feel free to write back if I can help further.
Michael Weylandt
On Aug 6, 2011, at 4:35 PM, smajor sema...@ncsu.edu wrote:
On 06-Aug-11 19:37:49, Alexander Engelhardt wrote:
On 08/06/2011 09:23 PM, Jorge Ivan Velez wrote:
Hi Alex,
Try
require(MASS)
Loading required package: MASS
b- c(2039L, 2088L, 5966L, 2353L, 1966L, 2312L, 3305L, 2013L, 3376L,
+ 3363L, 3567L, 4798L, 2032L, 1699L, 3001L, 2329L, 3944L, 2568L,
Hi Eduardo
I have just realized that I did not my send previous message to r-help
You do not need to put it in a float but you must
create the graph in a chunk, all you need is the \includegraphic command
This may avoid float space problems and make it easier for many graphs.
with the
40 matches
Mail list logo