Re: [R] replayPlot in loop
[EMAIL PROTECTED] wrote: Hi, In order to make a movie-like animation of different graphs with replayPlot inside a fast loop: is there a way to avoid the appearance of some white stripes/streaks between the different calls to replayPlot? In principle it is device dependend and the devices are generally not designed to display movies. Also drawing of a plot might take a considerable amount of time. Uwe Ligges thanks in advance Johan VH __ 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
Re: [R] graph tick label size
FISCHER, Matthew wrote: Hi R-users, I'm running R under Unix and producing postscript output of graphs. Soemtimes, some tick axis labels disappear from my output. eg if I have a vector months - c(J,F,M,A,M,J,J,A,S,O,N,D) Then the M and one or 2 other letters are dropped from the axis. This seems to be a size problem, but the spacing looks fine to me (all the letters could easily be seen if they were all there). So my question is, can I override what R is doing to the tick labels, without making the labels smaller? I searched the help files but couldn't find anything specifically on this. Not directly, but you can fake by plotting the stuff in two chunks as in: plot(1:12, xaxt=n) months - c(J,F,M,A,M,J,J,A,S,O,N,D) temp - seq(1, 12, 2) axis(1, temp, labels=months[temp]) temp - seq(2, 12, 2) axis(1, temp, labels=months[temp]) Uwe Ligges cheers, Matt. Dr Matt Fischer ANSTO - Institute for Nuclear Geophysiology PMB 1 Menai NSW 2234 Ph: +61 2 9717 9686 Fax: +61 2 9717 3599 Mobile: 0428 363 146 http://www.ansto.gov.au/nugeo/ http://ipilps.ansto.gov.au/ [[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-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] minimal hedge variance ratio
Hi all i have two data sets, spot and futures cash market prices. to estimate the minimum variance hedge ratio, i first had a glance on the correlation coefficient of relative price change (ln(St / St-1). surprizingly the value is just 0.2 compared to actual price correlation of 0.9. (i did regress the spot change on future change, co-effi is 0.3, and R2 is only 0.025 a) in such scenario can someone help me in estimating the ratio which are time varying. b) is there a way to define the function as the correlation will work at given level of basis (futures - spot). thank u for the help and co-operation rgds snvk __ 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] How to update R from ver 1.9.1 to 2
See the rw-FAQ ..., Q2.8 in the latest version. On Sat, 17 Sep 2005, Srinivas Iyyer wrote: Dear group, apologies if this is a stupid question. I searched CRAN sites. I am afraid I missed it. Can any one help me if I can update my windows version of 1.9.1 to 2 or higer. -- 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
[R] indicator value in labdsv
Hi, I'm trying to find out what threshold of indicator value in labadsv should be used to accept a specie as an indicator one? So far I assumed that indval=0.5 is high enough to avoid any mistakes but it was based only in my intuition. I'd be greatful for any advise best regards Agnieszka __ 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] distance to eye in persp()
Hi the manpage for persp() has a wonderful section where a the trans3d () function is used with points() and lines() to add red dots and a green sinusoid to the Mexican hat surface. Does anyone have a way to tell what distance a point is from the eye/ camera? Take the following line: lines (trans3d(x, y=10, z= 6 + sin(x), pm = res), col = 3) Is there a function like trans3d() that returns a vector of distances from the x,y,z point to the camera? I want this so I can plot clouds of points with the further ones in smaller plotsizes, and perhaps even fading to white (as though viewed through fog). -- Robin Hankin Uncertainty Analyst National Oceanography Centre, Southampton European Way, Southampton SO14 3ZH, UK tel 023-8059-7743 __ 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] dynamic object names?
I am trying to extract data from a matrix. Let's say that i am interested in extracting rows from a 4x4 matrix. Instead of giving a fix name to these 4 rows I would like to add a number to prefix. As result I should get 4 objects named: prefix_1 prefix_2 prefix_3 prefix_4 I attepted to solve the problem with a loop, but without success. Any hints?? matrix(LETTERS[1:16], ncol=4) - MM MM [,1] [,2] [,3] [,4] [1,] A E I M [2,] B F J N [3,] C G K O [4,] D H L P for (xxx in 1:4) { MM[xxx,] - prefix_xxx; } Thanks samir __ 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] dynamic object names?
[EMAIL PROTECTED] wrote: I am trying to extract data from a matrix. Let's say that i am interested in extracting rows from a 4x4 matrix. Instead of giving a fix name to these 4 rows I would like to add a number to prefix. As result I should get 4 objects named: prefix_1 prefix_2 prefix_3 prefix_4 I attepted to solve the problem with a loop, but without success. Any hints?? 1. In fact, you do not really want it. 2. You want to use - rather than - for assigments. Reading your code is really hard when you are using this non-conventional way. 3. In order to answer your question: This is the FAQ How can I turn a string into a variable?. The posting guide asks you to read these FAQs before sending questions to R-help. Please do so. Uwe Ligges matrix(LETTERS[1:16], ncol=4) - MM MM [,1] [,2] [,3] [,4] [1,] A E I M [2,] B F J N [3,] C G K O [4,] D H L P for (xxx in 1:4) { MM[xxx,] - prefix_xxx; } Thanks samir __ 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
Re: [R] distance to eye in persp()
On 19-Sep-05 Robin Hankin wrote: Hi the manpage for persp() has a wonderful section where a the trans3d() function is used with points() and lines() to add red dots and a green sinusoid to the Mexican hat surface. Does anyone have a way to tell what distance a point is from the eye/camera? Take the following line: lines (trans3d(x, y=10, z= 6 + sin(x), pm = res), col = 3) Is there a function like trans3d() that returns a vector of distances from the x,y,z point to the camera? I want this so I can plot clouds of points with the further ones in smaller plotsizes, and perhaps even fading to white (as though viewed through fog). Wonderfully put! That's what statistics is about! I think you may have to write your own. This is possible given the values for the parameters xlim, ylim, zlim, r, theta, phi (default as defined in ?persp, or explicitly user-defined), since you can then determine the 3D coordinates of the Eye relative to the (X,Y,Z) axes being plotted, after which the distance to a particular (x,y,z) point is trivial. E.g. 1. Coordinates of Eye relative to the centre of the box xE - r*sin(theta + pi)*cos(phi) yE - r*cos(theta + pi)*cos(phi) zE - r*sin(phi) 2. Centre of box relative to real (0,0,0) xC - mean(xlim); yC -mean(ylim); xC - mean(zlim) 3. Coordinates of (x,y,z) relative to Eye x1 - x - xE - xC; y1 - y - yE - yC; z1 - z - zE - zC 4. Distance from Eye to (x,y,z) d = sqrt(x1^2 + y1^2 + z1^2) (Hoping I've not got anything the wrong way round there!) Best wishes, Ted. E-Mail: (Ted Harding) [EMAIL PROTECTED] Fax-to-email: +44 (0)870 094 0861 Date: 19-Sep-05 Time: 10:17:55 -- 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
[R] Problem with tick marks in lines.survfit (package survival)
I have attempted to follow posting guidelines but I have failed to find out what I am doing wrong here. I am trying to use lines.survfit to plot a second curve onto a survival curve produced by plot.survfit. In my case this is to be a progression free survival curve superimposed upon an overall survival curve, but I will illustrate my problem using the example given in the help for lines.survfit. I would like to have tick marks for censored data points on both curves, so I would like to do something like: fit - survfit(Surv(time, status) ~ sex, pbc,subset=1:312) plot(fit[2], mark.time=TRUE, xscale=365.24, + xlab='Years', ylab='Survival') lines(fit[1], lwd=2, xscale=365.24, mark.time=TRUE) but when I do this, no tick marks appear on the second curve. Can anyone see what I am doing wrong here? Rachel [[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
Re: [R] Integrate functions with loops
Thanks Sundar what you suggested worked fine You will see that x is a vector and tt[i]^x returns a vector of the same length. You are trying to place this vector into a[i] which is length 1. Try the following *untested* code instead: untested integrandtotest - function(x) { sum(sapply(x, function(xi) sum(tt^xi))) } except you are summing twice so it should be. integrandtotest - function(x) {(sapply(x, function(xi) sum(tt^xi))) Alan Brennan Director of Health Economics and Decision Science http://www.shef.ac.uk/scharr/sections/heds ScHARR School of Health and Related Research University of Sheffield Regent Ct 30 Regent St Sheffield S1 4DA Tel:+44 (0)114 2220684 Fax:+44 (0)114 2724095 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
Re: [R] Integrate functions with loops
Thankyou Brian integrandtotest - function(x) colSums(outer(t, x, ^)) This worked fine in fact this integration can be done analytically (it is a sum of exponentials). I know but i have another five terms in the integrand as well as the one that was causing the trouble Many, many thanks Alan __ 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] Use of least-squares means, are they misleading?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi. Sorry for sending this question twice, but I would really like to know your opinion on this topic. I have been reading about the (in)convenience of using least-squares means (a. k. a. adjusted means) in multiple comparisons (I used to resort to them when using SAS). I even read a post in this list warning against them, but not giving much detail. What is your opinion? Should I avoid using LSmeans for comparison (e.g. after ANOVA)? Greetings. Felipe __ 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 -BEGIN PGP SIGNATURE- iEYEARECAAYFAkMulV4ACgkQWtdQtNzjBl6QpwCfXYweOQWipAYY2ZZFiDYuAiwJ LQYAnicmypM4yZhzMztRmZvppTZWes1z =696Z -END PGP SIGNATURE- __ 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] dynamic object names?
On Mon, 2005-09-19 at 11:38 +0200, Uwe Ligges wrote: [EMAIL PROTECTED] wrote: I am trying to extract data from a matrix. Let's say that i am interested in extracting rows from a 4x4 matrix. Instead of giving a fix name to these 4 rows I would like to add a number to prefix. As result I should get 4 objects named: prefix_1 prefix_2 prefix_3 prefix_4 Samir, In addition to the other comments, part of the problem may be that you can't assign the character '_' to standard objects. Could you use '.' instead? JC I attepted to solve the problem with a loop, but without success. Any hints?? 1. In fact, you do not really want it. 2. You want to use - rather than - for assigments. Reading your code is really hard when you are using this non-conventional way. 3. In order to answer your question: This is the FAQ How can I turn a string into a variable?. The posting guide asks you to read these FAQs before sending questions to R-help. Please do so. Uwe Ligges matrix(LETTERS[1:16], ncol=4) - MM MM [,1] [,2] [,3] [,4] [1,] A E I M [2,] B F J N [3,] C G K O [4,] D H L P for (xxx in 1:4) { MM[xxx,] - prefix_xxx; } Thanks samir __ 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-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] minimal hedge variance ratio
Krishna a écrit : i have two data sets, spot and futures cash market prices. to estimate the minimum variance hedge ratio, i first had a glance on the correlation coefficient of relative price change (ln(St / St-1). surprizingly the value is just 0.2 compared to actual price correlation of 0.9. The correlation coefficient measures the strengh of *linear* relation between 2 variables X and Y. Thus when you replace X by X'=aX+b or Y by Y'=cY+d (for instance use f = function(U) {return((U - mean(U))/sd(U));}) then you always have cor(X,Y)=cor(X',Y')=cor(X,Y')=cor(X',Y). But when you use X'=f(X), Y'=g(Y) with f,g non linear functions there is no reason to have cor(X',Y')=cor(X,Y). ... and f(X) = logreturn(X) is not really a linear transformation. hih __ 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] dynamic object names?
see ?paste hih __ 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] dynamic object names?
John Charles Considine wrote: On Mon, 2005-09-19 at 11:38 +0200, Uwe Ligges wrote: [EMAIL PROTECTED] wrote: I am trying to extract data from a matrix. Let's say that i am interested in extracting rows from a 4x4 matrix. Instead of giving a fix name to these 4 rows I would like to add a number to prefix. As result I should get 4 objects named: prefix_1 prefix_2 prefix_3 prefix_4 Samir, In addition to the other comments, part of the problem may be that you can't assign the character '_' to standard objects. Why not??? R can handle standard names with _ in it for ages now. Uwe Ligges Could you use '.' instead? JC I attepted to solve the problem with a loop, but without success. Any hints?? 1. In fact, you do not really want it. 2. You want to use - rather than - for assigments. Reading your code is really hard when you are using this non-conventional way. 3. In order to answer your question: This is the FAQ How can I turn a string into a variable?. The posting guide asks you to read these FAQs before sending questions to R-help. Please do so. Uwe Ligges matrix(LETTERS[1:16], ncol=4) - MM MM [,1] [,2] [,3] [,4] [1,] A E I M [2,] B F J N [3,] C G K O [4,] D H L P for (xxx in 1:4) { MM[xxx,] - prefix_xxx; } Thanks samir __ 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-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
Re: [R] month increment for chron dates
Gabor, This is great. Thank you. It runs in just 3 seconds. Brian __ 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] distance to eye in persp()
(Ted Harding) wrote: On 19-Sep-05 Robin Hankin wrote: Hi the manpage for persp() has a wonderful section where a the trans3d() function is used with points() and lines() to add red dots and a green sinusoid to the Mexican hat surface. Does anyone have a way to tell what distance a point is from the eye/camera? Take the following line: lines (trans3d(x, y=10, z= 6 + sin(x), pm = res), col = 3) Is there a function like trans3d() that returns a vector of distances from the x,y,z point to the camera? I want this so I can plot clouds of points with the further ones in smaller plotsizes, and perhaps even fading to white (as though viewed through fog). Wonderfully put! That's what statistics is about! I think you may have to write your own. This is possible given the values for the parameters xlim, ylim, zlim, r, theta, phi (default as defined in ?persp, or explicitly user-defined), since you can then determine the 3D coordinates of the Eye relative to the (X,Y,Z) axes being plotted, after which the distance to a particular (x,y,z) point is trivial. E.g. 1. Coordinates of Eye relative to the centre of the box xE - r*sin(theta + pi)*cos(phi) yE - r*cos(theta + pi)*cos(phi) zE - r*sin(phi) 2. Centre of box relative to real (0,0,0) xC - mean(xlim); yC -mean(ylim); xC - mean(zlim) 3. Coordinates of (x,y,z) relative to Eye x1 - x - xE - xC; y1 - y - yE - yC; z1 - z - zE - zC 4. Distance from Eye to (x,y,z) d = sqrt(x1^2 + y1^2 + z1^2) (Hoping I've not got anything the wrong way round there!) I think you forgot the rescaling to [0,1], but it's probably even easier than that. persp returns the 4x4 transformation matrix used to take user coordinates into screen coordinates. trans3d uses that matrix to extract the screen x and screen y coordinates (extend the 3-vector to homogeneous coordinates by appending a 1, multiply by the projection matrix, convert back to Euclidean coordinates by dividing by the last coordinate). You can probably just do what trans3d does, but keep the z-coordinate, to get a reasonable measure of depth. i.e. depth3d - function(x,y,z, pmat) { tr - cbind(x, y, z, 1) %*% pmat return(tr[,3]/tr[,4]) } This is sufficient for doing fog calculations. For perspective shrinkage, you'll need to say where the user's eye is in these coordinates (or just use depths as distances). Duncan Murdoch __ 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] Converting the result of classification tree into SQL query
I built the model of classification tree.Now I want to convert the result rules of the tree into Sql query for applying received rules to my dataset in the database managment system. Can this be done or not ? If it can, tell me please how I can do it. Thanks for your help. __ 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] Teaching R - In front of the computer?
Dear R-Users, given you have been teaching R to students (grad level, mainly social science background, no previous programming experience, 80% know SPSS), what are your experiences concerning the style of teaching? Do you prefer to stand in front of the class like in normal lectures and you show them slides? Or do you you explain some concept (for example things like mydata[order(var1, var2),]) and show it directly on the computer via beamer/projector and also the students have to enter it on the computers in front of them. Any experiences you can share are highly appreciated. Thanks, Roland + This mail has been sent through the MPI for Demographic Rese...{{dropped}} __ 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] toLatex
hallo! does everybody know which kind of objects the function toLatex can manage? thanks simone vantini -- Simone Vantini MOX (Modelling and Scientific Computing) Dipartimento di Matematica F. Brioschi Politecnico di Milano P.za Leonardo da Vinci, 32 20133 Milano (Italy) tel: +39 02 2399 4604 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
[R] Converting the result of classification tree into SQL query
I built the model of classification tree.Now I want to convert the result rules of the tree into Sql query for applying received rules to my dataset in the database managment system. Can this be done or not ? If it can, tell me please how I can do it. Thanks for your help. Yulia __ 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] toLatex
Try 'methods(toLatex)'. -roger Simone Vantini wrote: hallo! does everybody know which kind of objects the function toLatex can manage? thanks simone vantini -- Roger D. Peng http://www.biostat.jhsph.edu/~rpeng/ __ 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] Teaching R - In front of the computer?
Hi the only way I have ever made this work is to have an R session projected on the screen, and each student to have their own machine on which they can try out what I type. You can then set the students questions to force them to play around with R syntax. A good one is to put up a graph that you have prepared beforehand, and say reproduce this. Just be careful with ad-hoc playing around, in a live public session. I got badly bitten by this: f - function(x){(x*x -1) - (x-1)*(x+1) } f(10:100010) [everything behaving as documented, but embarrassingly unexpected!] best Robin On 19 Sep 2005, at 14:25, Rau, Roland wrote: Dear R-Users, given you have been teaching R to students (grad level, mainly social science background, no previous programming experience, 80% know SPSS), what are your experiences concerning the style of teaching? Do you prefer to stand in front of the class like in normal lectures and you show them slides? Or do you you explain some concept (for example things like mydata[order(var1, var2),]) and show it directly on the computer via beamer/projector and also the students have to enter it on the computers in front of them. Any experiences you can share are highly appreciated. Thanks, Roland + This mail has been sent through the MPI for Demographic Rese... {{dropped}} __ 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 -- Robin Hankin Uncertainty Analyst National Oceanography Centre, Southampton European Way, Southampton SO14 3ZH, UK tel 023-8059-7743 __ 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] R(D)COM and Excel compile error
Dear R-users, I tried to install the R(D)COM server version 1.35, to connect MS Excel and R. I am using R-version 2.1.0 on a Windows environment. And I am using MS Excel 2000 in a dutch version. I get the following error when I open Excel: Compile error, can not find the project or the library (translated from dutch). And it indicates LEFT in blue bold. Does anyone of you know what could have caused this error? Does the R(D)COM only work with English Excel versions? Thanks, Jan Jan Wijffels University Center for Statistics W. de Croylaan 54 3001 Heverlee Belgium tel: +32 (0)16 322784 fax: +32 (0)16 322831 http://www.kuleuven.be/ucs http://www.kuleuven.be/ucs Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm [[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
Re: [R] Converting the result of classification tree into SQL query
On 9/19/05 9:32 AM, Yuliya S Ivanova [EMAIL PROTECTED] wrote: I built the model of classification tree.Now I want to convert the result rules of the tree into Sql query for applying received rules to my dataset in the database managment system. Can this be done or not ? If it can, tell me please how I can do it. Thanks for your help. This will need to be done by hand, I think. In other words, you will need to write the SQL query yourself. Of course, you can make such a query dynamic using tools like paste, but I don't think there is a tool to do exactly what you want. Sean __ 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] Use of least-squares means, are they misleading?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Thank you, I will try there. Felipe Peter Flom wrote: | Felipe | | I doubt you will get much response on this list, although it's an | interesting question. You might try asking it on another list. Some | that often have discussions of such topics are | | SAS-L | EDSTAT-L and | STAT-L | | Regards | | Peter | | Peter L. Flom, PhD | Assistant Director, Statistics and Data Analysis Core | Center for Drug Use and HIV Research | National Development and Research Institutes | 71 W. 23rd St | http://cduhr.ndri.org | www.peterflom.com | New York, NY 10010 | (212) 845-4485 (voice) | (917) 438-0894 (fax) | | | |Felipe [EMAIL PROTECTED] 09/19/05 6:39 AM | | Hi. | Sorry for sending this question twice, but I would really like to know | your opinion on this topic. | | I have been reading about the (in)convenience of using least-squares | means (a. k. a. adjusted means) in multiple comparisons (I used to | resort to them when using SAS). I even read a post in this list warning | against them, but not giving much detail. | | What is your opinion? Should I avoid using LSmeans for comparison (e.g. | after ANOVA)? | | Greetings. | | Felipe | | __ | 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 -BEGIN PGP SIGNATURE- iEYEARECAAYFAkMuxT0ACgkQWtdQtNzjBl7LWACcCddjPgTnZX2+15aDLbxe+qxD rHcAoIdkgW8ke/M8bb7pSlcptTz9K53j =eCEC -END PGP SIGNATURE- __ 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] Problem with tick marks in lines.survfit (package survival)
How about: plot(fit, mark.time=TRUE, xscale=365.24,xlab='Years', ylab='Survival',lwd=c(2,1)) Rob - Original Message - From: Rachel Pearce [EMAIL PROTECTED] To: r-help@stat.math.ethz.ch Sent: Monday, September 19, 2005 4:59 AM Subject: [R] Problem with tick marks in lines.survfit (package survival) I have attempted to follow posting guidelines but I have failed to find out what I am doing wrong here. I am trying to use lines.survfit to plot a second curve onto a survival curve produced by plot.survfit. In my case this is to be a progression free survival curve superimposed upon an overall survival curve, but I will illustrate my problem using the example given in the help for lines.survfit. I would like to have tick marks for censored data points on both curves, so I would like to do something like: fit - survfit(Surv(time, status) ~ sex, pbc,subset=1:312) plot(fit[2], mark.time=TRUE, xscale=365.24, + xlab='Years', ylab='Survival') lines(fit[1], lwd=2, xscale=365.24, mark.time=TRUE) but when I do this, no tick marks appear on the second curve. Can anyone see what I am doing wrong here? Rachel [[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-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] Teaching R - In front of the computer?
Rau, Roland wrote: Dear R-Users, given you have been teaching R to students (grad level, mainly social science background, no previous programming experience, 80% know SPSS), what are your experiences concerning the style of teaching? Do you prefer to stand in front of the class like in normal lectures and you show them slides? Or do you you explain some concept (for example things like mydata[order(var1, var2),]) and show it directly on the computer via beamer/projector and also the students have to enter it on the computers in front of them. Any experiences you can share are highly appreciated. I like to teach using a projector and people sitting in front of other machines beeing able to try out some of the nonsense I am telling. Of course, they need more exercises than just the few minutes in between my sentences. I am always noth using slides and showing the examples in R directly. For all other stuff, it is the same as with other courses and lectures. If this is a one-week course, I'd like to propose 1.5 hours course + 1.5 hours exercise in the morning and the same again in the afternoon. If this is a whole-term course, you may be able to let people do homework, but always talk about their results and show them hoe to do it in a good fashion. Unfortunately, for courses with lots of poeple (say 90, for the first-years' R course I am teaching) it is impossible that people are sitting in front of a computer for some reasons: not enough machines, no big rooms, and too much noise caused by both machines and people. The really important fact from my point of view: Give people many *exercises* and the *time* to work on these exercises. Expect that people without any programming experience will be much slower than you imagine in your worst nightmares. ;-) Uwe Ligges Thanks, Roland + This mail has been sent through the MPI for Demographic Rese...{{dropped}} __ 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
Re: [R] distance to eye in persp()
Duncan You can probably just do what trans3d does, but keep the z-coordinate, to get a reasonable measure of depth. i.e. depth3d - function(x,y,z, pmat) { tr - cbind(x, y, z, 1) %*% pmat return(tr[,3]/tr[,4]) } fantastic! This is exactly what I was looking for. This is sufficient for doing fog calculations. For perspective shrinkage, you'll need to say where the user's eye is in these coordinates (or just use depths as distances). I was going to use hsv(h=1, v=1, s=exp(d/d0) ) (some people believe that red light [ie hsv(h=1, )] penetrates further through mist than other colours. I can now test this !-) The other idea I had was to plot points with less than the median depth with big symbols and the others with small symbols. That would seem to be a nice nonparametric way to do it. Any other ideas? very best wishes to everyone Robin -- Robin Hankin Uncertainty Analyst National Oceanography Centre, Southampton European Way, Southampton SO14 3ZH, UK tel 023-8059-7743 __ 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] Problem with tick marks in lines.survfit (package survival)
Sorry, yes I think that would work, for the example I gave, but I am trying to superimpose a curve onto an existing plot. The option of using plot to draw both curves is not available, since the data are (or were) in two different columns. For that reason I was trying to use lines for the second curve. I used the example given with the package so that I didn't have to publish my own data (which are not mine to publish, if you see what I mean). I have now given up and created a new data frame with OS and PFS stacked on top of each other, but this is a very time-consuming workaround. Do you (or anyone) know how to make lines.survfit draw tick marks? Rachel On 19/09/05, Robert Baer [EMAIL PROTECTED] wrote: How about: plot(fit, mark.time=TRUE, xscale=365.24,xlab='Years', ylab='Survival',lwd=c(2,1)) Rob - Original Message - From: Rachel Pearce [EMAIL PROTECTED] To: r-help@stat.math.ethz.ch Sent: Monday, September 19, 2005 4:59 AM Subject: [R] Problem with tick marks in lines.survfit (package survival) I have attempted to follow posting guidelines but I have failed to find out what I am doing wrong here. I am trying to use lines.survfit to plot a second curve onto a survival curve produced by plot.survfit. In my case this is to be a progression free survival curve superimposed upon an overall survival curve, but I will illustrate my problem using the example given in the help for lines.survfit. I would like to have tick marks for censored data points on both curves, so I would like to do something like: fit - survfit(Surv(time, status) ~ sex, pbc,subset=1:312) plot(fit[2], mark.time=TRUE, xscale=365.24, + xlab='Years', ylab='Survival') lines(fit[1], lwd=2, xscale=365.24, mark.time=TRUE) but when I do this, no tick marks appear on the second curve. Can anyone see what I am doing wrong here? Rachel [[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-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] Change in behavior of compare statement
Dear all, I have come across some Windows Script code which calls the file.exists function in R to check for the existence of a particular file on an R server. This is what the code looks like: do { // some useful code } while (m_workspace.session.eval(file.exists( + dataFile + )) != 1); It appears that the behavior of this code is different when used with R version 2.1.1 compared to R version 1.9.1. If the compare statement s is changed from while (s != 1) to while (s != True) or while !(s) the behavior of the code is the same with both versions of R. After reading the R documentation, I have not found any evidence that the original syntax (s != 1) is in any way recommended or supported (in version 2.1.1). Still, my questions are: 1. Is it possible that the syntax has been supported in older versions of R? 2. Is the reason for the recent change in behavior known and documented anywhere? I have been unable to find an answer to either question, so any information about this would be appreciated. Thanks, T. Andersson [[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] FDR analyses: minimum number of features
Dear List, We are planning a genotyping study to be analyzed using false discovery rates (FDRs) (See Storey and Tibshirani PNAS 2003; 100:9440-5). I am interested in learning if there is any consensus as to how many features (ie. how many P values) need to be studied before reasonably reliable FDRs can be derived. Does anyone know of a citation where this is discussed? Bill Dupont William D. Dupont phone: 615-343-4100 URL http://biostat.mc.vanderbilt.edu/twiki/bin/view/Main/WilliamDupont __ 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] indicator value in labdsv
Agnieszka, As Jari indicated, it depends on which function you meant in you inquiry. The duleg() function implements the Dufrene-Legendre algorithm, where indicator species are indicative of a priori communities. It this requires a classification, and is biased to find species which occur in the dataset approximately as often as the mean cluster size. The indpsc() function calculates the mean similarity of all samples a species occurs in. This is slightly biased because we know that the samples being used to calculate the mean share at least the species that defines them, but it is still possible to compare those values to the mean similarity of the whole matrix, or to an expectation of maximum similarity. Obviously, as species occur more frequently, the harder it is to have a really high similarity (indicator value), with the extreme case that a species that occurs in every sample must have the same value as the mean of the whole matrix. To tell the truth, I forgot that indspc() was included in the current version of labdsv. In the new version (due to be released any day), I have included a permutation test that estimates quantiles of expected values for different numbers of occurrences. It works, but is pretty slow. Jari has created a version that uses parametric statistics to estimate the same envelope, but I haven't had a chance to try it yet. What research are you doing, and what are you really trying to determine? Perhaps something altogether different will work better. Thanks, Dave Roberts On Mon, 2005-09-19 at 09:41 +0200, [EMAIL PROTECTED] wrote: Hi, I'm trying to find out what threshold of indicator value in labadsv should be used to accept a specie as an indicator one? So far I assumed that indval=0.5 is high enough to avoid any mistakes but it was based only in my intuition. I'd be greatful for any advise best regards Agnieszka, R mailing list software appends the following to your message: PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Then about indicator value analysis. You should be more specific: there seem to be three alternatives functions for indicator species in labdsv. Which did you mean? At least two of these return an item called indval, and these two alternative indvals are very different. For the Dufrêne-Legendre indvals, you should check the original paper (see references in the help page), and there you even have an associated P value. In indspc, the variance of the indval clearly is dependent on species frequency. Moreover, in indspc the expected indval (and its variance) are dependent on the whole set of sites you have: these reflect the general homogeneity of your data set. Therefore you cannot say there that any certain value would mean that a species is a good indicator. However, it would be easy to work out standard errors for indspc indvals. I think it would be more useful to post to some other mailing group where people are more concerned about indicator species, or to contact the package author directly (I CC this message to him). cheers, jari oksanen -- David W. Roberts office 406-994-4548 Professor and Head FAX 406-994-3190 Department of Ecology email [EMAIL PROTECTED] Montana State University Bozeman, MT 59717-3460 __ 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] indicator value in labdsv
Hello, I was uclear before, I'm sory about it. I forgot to add that I'm using duleg... I used mvpart for multivariate regression trees. My input variables are environmental parameters, output variables are macrophyte species (presence=1,absence=0 in conecutive cases=lakes). For obtained classes I used duleg to find indicator species for every class. I checked the article Dufrene, M. and Legendre, P. 1997. Species assemblages and indicator species: the need for a flexible asymmetrical approach. Ecol. Monogr. 67(3):345-366. The authors used the threshold of indval=0.25(25%) and that's the only hint I've found in the literature. This threshod seems to reasonable, but still I have impression that's too low... best regards Agnieszka Agnieszka, As Jari indicated, it depends on which function you meant in you inquiry. The duleg() function implements the Dufrene-Legendre algorithm, where indicator species are indicative of a priori communities. It this requires a classification, and is biased to find species which occur in the dataset approximately as often as the mean cluster size. The indpsc() function calculates the mean similarity of all samples a species occurs in. This is slightly biased because we know that the samples being used to calculate the mean share at least the species that defines them, but it is still possible to compare those values to the mean similarity of the whole matrix, or to an expectation of maximum similarity. Obviously, as species occur more frequently, the harder it is to have a really high similarity (indicator value), with the extreme case that a species that occurs in every sample must have the same value as the mean of the whole matrix. To tell the truth, I forgot that indspc() was included in the current version of labdsv. In the new version (due to be released any day), I have included a permutation test that estimates quantiles of expected values for different numbers of occurrences. It works, but is pretty slow. Jari has created a version that uses parametric statistics to estimate the same envelope, but I haven't had a chance to try it yet. What research are you doing, and what are you really trying to determine? Perhaps something altogether different will work better. Thanks, Dave Roberts On Mon, 2005-09-19 at 09:41 +0200, [EMAIL PROTECTED] wrote: Hi, I'm trying to find out what threshold of indicator value in labadsv should be used to accept a specie as an indicator one? So far I assumed that indval=0.5 is high enough to avoid any mistakes but it was based only in my intuition. I'd be greatful for any advise best regards Agnieszka, R mailing list software appends the following to your message: PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Then about indicator value analysis. You should be more specific: there seem to be three alternatives functions for indicator species in labdsv. Which did you mean? At least two of these return an item called indval, and these two alternative indvals are very different. For the Dufręne-Legendre indvals, you should check the original paper (see references in the help page), and there you even have an associated P value. In indspc, the variance of the indval clearly is dependent on species frequency. Moreover, in indspc the expected indval (and its variance) are dependent on the whole set of sites you have: these reflect the general homogeneity of your data set. Therefore you cannot say there that any certain value would mean that a species is a good indicator. However, it would be easy to work out standard errors for indspc indvals. I think it would be more useful to post to some other mailing group where people are more concerned about indicator species, or to contact the package author directly (I CC this message to him). cheers, jari oksanen -- Best regards, mailto:[EMAIL PROTECTED] Agnieszka Strzelczak, Research Assistant mailto:[EMAIL PROTECTED] Institute of Chemistry and Environmental Protection Faculty of Chemical Engineering Szczecin University of Technology Aleja Piastow 42 71-065 Szczecin Poland __ 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] Change in behavior of compare statement
Dear all, I have come across some Windows Script code which calls the file.exists function in R to check for the existence of a particular file on an R server. This is what the code looks like: do { // some useful code } while (m_workspace.session.eval(file.exists( + dataFile + )) != 1); It appears that the behavior of this code is different when used with R version 2.1.1 compared to R version 1.9.1. If the compare statement s is changed from while (s != 1) to while (s != True) or while !(s) the behavior of the code is the same with both versions of R. After reading the R documentation, I have not found any evidence that the original syntax (s != 1) is in any way recommended or supported (in version 2.1.1). Still, my questions are: 1. Is it possible that the syntax has been supported in older versions of R? 2. Is the reason for the recent change in behavior known and documented anywhere? I have been unable to find an answer to either question, so any information about this would be appreciated. Thanks, T. Andersson __ 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] How to mimic pdMat of lme under lmer?
Dear members, I would like to switch from nlme to lme4 and try to translate some of my models that worked fine with lme. I have problems with the pdMat classes. Below a toy dataset with a fixed effect F and a random effect R. I gave also 2 similar lme models. The one containing pdLogChol (lme1) is easy to translate (as it is an explicit notation of the default model) The more parsimonious model with pdDiag replacing pdLogChol I cannot reproduce with lmer. The obvious choice for me would be my model lmer2, but this is yielding different result. Somebody any idea? Thanks, Joris I am using R version 2.1.0 for Linux and the most recent downloads of Matrix and nlme #dataset from McLean, Sanders and Stroup F - factor(c(1,1,1,1,1,1,2,2,2,2,2,2)) R - factor(c(1,1,2,2,3,3,1,1,2,2,3,3)) s - c(51.43,51.28,50.93,50.75,50.47,50.83,51.91,52.43,52.26,52.33,51.58,51.23) DS - data.frame(F,R,s) DS$F - as.factor(DS$F) DS$R - as.factor(DS$R) library(nlme) lme1 - lme(data = DS,s ~ F,random = list(R = pdLogChol(~F))) lme2 - lme(data = DS,s ~ F,random = list(R = pdDiag(~F))) summary(lme1) summary(lme2) library(lme4) lmer1 - lmer(data = DS,s ~ F + (F|R)) lmer2 - lmer(data = DS,s ~ F + (1|R) + (1|F)) summary(lmer1) summary(lmer2) confidentiality notice: The information contained in this e-mail is confidential and...{{dropped}} __ 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] anova for random-intercept lmer
Hi, This might be a silly question, but how do I create a pooled model (binomial family) as a baseline for a likelihood-ratio test for a random-intercept lmer model? R won't compare a glm model with an lmer model (or glmmPQL ). Thanks in advance. -Bobby __ 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] indicator value in labdsv
Wow! That was fast! Unfortunately, Agnieszka, I don't think you will find an objective criterion for this. Clearly, species which do not have a statistically significant value are probably less useful, but of the many that are significant, many may be marginal. Without knowing fully what you are hoping to achieve, I think I would rank the species by indicator value, and establish the highest threshold for indicator value that gives you a suitable number of species for each type. That way, if you are looking to write a field key, for example, you would have sufficient values to identify every type I suspect. Good luck, Dave [EMAIL PROTECTED] wrote: Hello, I was uclear before, I'm sory about it. I forgot to add that I'm using duleg... I used mvpart for multivariate regression trees. My input variables are environmental parameters, output variables are macrophyte species (presence=1,absence=0 in conecutive cases=lakes). For obtained classes I used duleg to find indicator species for every class. I checked the article Dufrene, M. and Legendre, P. 1997. Species assemblages and indicator species: the need for a flexible asymmetrical approach. Ecol. Monogr. 67(3):345-366. The authors used the threshold of indval=0.25(25%) and that's the only hint I've found in the literature. This threshod seems to reasonable, but still I have impression that's too low... best regards Agnieszka Agnieszka, As Jari indicated, it depends on which function you meant in you inquiry. The duleg() function implements the Dufrene-Legendre algorithm, where indicator species are indicative of a priori communities. It this requires a classification, and is biased to find species which occur in the dataset approximately as often as the mean cluster size. The indpsc() function calculates the mean similarity of all samples a species occurs in. This is slightly biased because we know that the samples being used to calculate the mean share at least the species that defines them, but it is still possible to compare those values to the mean similarity of the whole matrix, or to an expectation of maximum similarity. Obviously, as species occur more frequently, the harder it is to have a really high similarity (indicator value), with the extreme case that a species that occurs in every sample must have the same value as the mean of the whole matrix. To tell the truth, I forgot that indspc() was included in the current version of labdsv. In the new version (due to be released any day), I have included a permutation test that estimates quantiles of expected values for different numbers of occurrences. It works, but is pretty slow. Jari has created a version that uses parametric statistics to estimate the same envelope, but I haven't had a chance to try it yet. What research are you doing, and what are you really trying to determine? Perhaps something altogether different will work better. Thanks, Dave Roberts On Mon, 2005-09-19 at 09:41 +0200, [EMAIL PROTECTED] wrote: Hi, I'm trying to find out what threshold of indicator value in labadsv should be used to accept a specie as an indicator one? So far I assumed that indval=0.5 is high enough to avoid any mistakes but it was based only in my intuition. I'd be greatful for any advise best regards Agnieszka, R mailing list software appends the following to your message: PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Then about indicator value analysis. You should be more specific: there seem to be three alternatives functions for indicator species in labdsv. Which did you mean? At least two of these return an item called indval, and these two alternative indvals are very different. For the Dufręne-Legendre indvals, you should check the original paper (see references in the help page), and there you even have an associated P value. In indspc, the variance of the indval clearly is dependent on species frequency. Moreover, in indspc the expected indval (and its variance) are dependent on the whole set of sites you have: these reflect the general homogeneity of your data set. Therefore you cannot say there that any certain value would mean that a species is a good indicator. However, it would be easy to work out standard errors for indspc indvals. I think it would be more useful to post to some other mailing group where people are more concerned about indicator species, or to contact the package author directly (I CC this message to him). cheers, jari oksanen -- Best regards, mailto:[EMAIL PROTECTED] Agnieszka Strzelczak, Research Assistant mailto:[EMAIL PROTECTED] Institute of Chemistry and Environmental Protection Faculty of Chemical Engineering Szczecin University of Technology Aleja Piastow 42 71-065 Szczecin Poland --
Re: [R] Teaching R - In front of the computer?
Adding to Uwe's comment, in my experience is also useful to use a text editor that connects to R (i.e. in Windows you have Tinn-R, jgr, SciViews) so people can see the function arguments as they type. People are accustomed to this feature from Excel so it helps them to fell more comfortable with the syntax. And yes, some students used to the mainstream point-and-click software can be REALLY slow so you may want to plan for that! Good luck! Francisco From: Uwe Ligges [EMAIL PROTECTED] To: Rau, Roland [EMAIL PROTECTED] CC: r-help@stat.math.ethz.ch Subject: Re: [R] Teaching R - In front of the computer? Date: Mon, 19 Sep 2005 16:14:57 +0200 Rau, Roland wrote: Dear R-Users, given you have been teaching R to students (grad level, mainly social science background, no previous programming experience, 80% know SPSS), what are your experiences concerning the style of teaching? Do you prefer to stand in front of the class like in normal lectures and you show them slides? Or do you you explain some concept (for example things like mydata[order(var1, var2),]) and show it directly on the computer via beamer/projector and also the students have to enter it on the computers in front of them. Any experiences you can share are highly appreciated. I like to teach using a projector and people sitting in front of other machines beeing able to try out some of the nonsense I am telling. Of course, they need more exercises than just the few minutes in between my sentences. I am always noth using slides and showing the examples in R directly. For all other stuff, it is the same as with other courses and lectures. If this is a one-week course, I'd like to propose 1.5 hours course + 1.5 hours exercise in the morning and the same again in the afternoon. If this is a whole-term course, you may be able to let people do homework, but always talk about their results and show them hoe to do it in a good fashion. Unfortunately, for courses with lots of poeple (say 90, for the first-years' R course I am teaching) it is impossible that people are sitting in front of a computer for some reasons: not enough machines, no big rooms, and too much noise caused by both machines and people. The really important fact from my point of view: Give people many *exercises* and the *time* to work on these exercises. Expect that people without any programming experience will be much slower than you imagine in your worst nightmares. ;-) Uwe Ligges Thanks, Roland + This mail has been sent through the MPI for Demographic Rese...{{dropped}} __ 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-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] indicator value in labdsv
That's what I was afraid of.. I'm not a biologist, so I have to involve my co-workes who are well-versed in biological issues. Thank you very, very much for help!!! best regards Agnieszka Monday, September 19, 2005, 6:38:35 PM, you wrote: Wow! That was fast! Unfortunately, Agnieszka, I don't think you will find an objective criterion for this. Clearly, species which do not have a statistically significant value are probably less useful, but of the many that are significant, many may be marginal. Without knowing fully what you are hoping to achieve, I think I would rank the species by indicator value, and establish the highest threshold for indicator value that gives you a suitable number of species for each type. That way, if you are looking to write a field key, for example, you would have sufficient values to identify every type I suspect. Good luck, Dave [EMAIL PROTECTED] wrote: Hello, I was uclear before, I'm sory about it. I forgot to add that I'm using duleg... I used mvpart for multivariate regression trees. My input variables are environmental parameters, output variables are macrophyte species (presence=1,absence=0 in conecutive cases=lakes). For obtained classes I used duleg to find indicator species for every class. I checked the article Dufrene, M. and Legendre, P. 1997. Species assemblages and indicator species: the need for a flexible asymmetrical approach. Ecol. Monogr. 67(3):345-366. The authors used the threshold of indval=0.25(25%) and that's the only hint I've found in the literature. This threshod seems to reasonable, but still I have impression that's too low... best regards Agnieszka Agnieszka, As Jari indicated, it depends on which function you meant in you inquiry. The duleg() function implements the Dufrene-Legendre algorithm, where indicator species are indicative of a priori communities. It this requires a classification, and is biased to find species which occur in the dataset approximately as often as the mean cluster size. The indpsc() function calculates the mean similarity of all samples a species occurs in. This is slightly biased because we know that the samples being used to calculate the mean share at least the species that defines them, but it is still possible to compare those values to the mean similarity of the whole matrix, or to an expectation of maximum similarity. Obviously, as species occur more frequently, the harder it is to have a really high similarity (indicator value), with the extreme case that a species that occurs in every sample must have the same value as the mean of the whole matrix. To tell the truth, I forgot that indspc() was included in the current version of labdsv. In the new version (due to be released any day), I have included a permutation test that estimates quantiles of expected values for different numbers of occurrences. It works, but is pretty slow. Jari has created a version that uses parametric statistics to estimate the same envelope, but I haven't had a chance to try it yet. What research are you doing, and what are you really trying to determine? Perhaps something altogether different will work better. Thanks, Dave Roberts On Mon, 2005-09-19 at 09:41 +0200, [EMAIL PROTECTED] wrote: Hi, I'm trying to find out what threshold of indicator value in labadsv should be used to accept a specie as an indicator one? So far I assumed that indval=0.5 is high enough to avoid any mistakes but it was based only in my intuition. I'd be greatful for any advise best regards Agnieszka, R mailing list software appends the following to your message: PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Then about indicator value analysis. You should be more specific: there seem to be three alternatives functions for indicator species in labdsv. Which did you mean? At least two of these return an item called indval, and these two alternative indvals are very different. For the Dufręne-Legendre indvals, you should check the original paper (see references in the help page), and there you even have an associated P value. In indspc, the variance of the indval clearly is dependent on species frequency. Moreover, in indspc the expected indval (and its variance) are dependent on the whole set of sites you have: these reflect the general homogeneity of your data set. Therefore you cannot say there that any certain value would mean that a species is a good indicator. However, it would be easy to work out standard errors for indspc indvals. I think it would be more useful to post to some other mailing group where people are more concerned about indicator species, or to contact the package author directly (I CC this message to him). cheers, jari oksanen __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide!
[R] factor as seq() in for loop
Dear all, I would like to use the values in vegaggr.BLMCMR02$colony str(vegaggr.BLMCMR02) `data.frame': 1678 obs. of 3 variables: $ vegtype : Factor w/ 27 levels 2010,2020,..: 3 4 5 19 4 5 19 5 $ colony: Factor w/ 406 levels 0,1,10,100,..: 1 1 1 1 2 2 2 $ Totvegproparea: num 0.00055 0.03956 0.95705 0.00284 0.05215 ... as the seq in a for loop (for(var in seq)), and therefore first picked them out as cols = aggregate(list(A = vegaggr.BLMCMR02$Totvegproparea), list(colony = vegaggr.BLMCMR02$colony), sum)[,1] str(cols) Factor w/ 406 levels 0,1,10,100,..: 1 2 3 4 5 6 7 8 9 10 ... I next planned to transform cols using as.integer(cols). However, this transformation gives a vector corresponding to seq(from=1,to=length(cols)). Could someone please give advice on how to make use of a factor as the seq in a for loop (which is strictly not allowed' according to ?Control). Thanks! Tord __ 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] Change in behavior of compare statement
On 9/19/2005 12:25 PM, Tomas Andersson wrote: Dear all, I have come across some Windows Script code which calls the file.exists function in R to check for the existence of a particular file on an R server. This is what the code looks like: do { // some useful code } while (m_workspace.session.eval(file.exists( + dataFile + )) != 1); What is this function m_workspace.session.eval? It's not a base function. Perhaps its definition has changed? It appears that the behavior of this code is different when used with R version 2.1.1 compared to R version 1.9.1. If the compare statement s is changed from while (s != 1) to while (s != True) or while !(s) the behavior of the code is the same with both versions of R. After reading the R documentation, I have not found any evidence that the original syntax (s != 1) is in any way recommended or supported (in version 2.1.1). Still, my questions are: s != 1 is perfectly valid R code. It will evaluate to TRUE in at least the following case: as.character(s) is not an NA, but isn't 1. s != True will evaluate to TRUE in most cases, unless s happens to contain that exact string. 1. Is it possible that the syntax has been supported in older versions of R? 2. Is the reason for the recent change in behavior known and documented anywhere? The syntax is fine, but it looks like a behaviour change in that function m_workspace.session.eval. You'll need to ask the author of that function what is happening. Duncan Murdoch __ 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] Teaching R - In front of the computer?
On 19-Sep-05 Rau, Roland wrote: Dear R-Users, given you have been teaching R to students (grad level, mainly social science background, no previous programming experience, 80% know SPSS),what are your experiences concerning the style of teaching? Do you prefer to stand in front of the class like in normal lectures and you show them slides? Or do you you explain some concept (for example things like mydata[order(var1, var2),]) and show it directly on the computer via beamer/projector and also the students have to enter it on the computers in front of them. Any experiences you can share are highly appreciated. Thanks, Roland I'd like to suggest (perhaps more for discussion than for implementation) thinking of the use of 'xmx' (X MultipleXer). See http://www.cs.brown.edu/software/xmx/ This work in X Windows on Unix/Linux. The essence of it is that the software is installed on each of a group of networked computers. One of these is the master, who starts it up in Floor mode and nominates any or all of the others to be in Floor, Seat or View mode (terminilogy apparently inspired by what happens at meetings). This causes an W window to appear on all nominated computers, within which is running a virutal X session. Anyone with Floor rights can do anything in that session, and it will be mirrored to all others. This includes keyboard input and mouse movements. Anyone in View mode an only see what is going on, but cannot make any input. In Seat mode, a user can virtually raise their hand and be granted Floor privileges. Anything anybody does outside the XMX window is private to them, and will not be seen by others, nor have any effect on the XMX session. This software seems not to have undergone further development in its public form for some time (apparently June 1999). I don't know if there have been further developments in a private or proprietary form. Its creator, John Bazik at Brown University, states that Xmx provides reasonable real-time performance on a local 10 Mb/sec ethernet on 52 UltraSparc-class machines. It might well provide a good vehicle for expository teaching in a class that large, since each student can see on their own screen what the instructor is displaying (and this could include a slide-show presentation alongside code files, commands being run in an R window, graphics plots, etc.). However, you would not want a class that size to all be in Floor mode! Seat mode could be OK, though, since anyone who wants to ask a question can go into Floor and then thsy can input things into the XMX session, or even just move their mouse around to indicate what they are talking about -- something which in a normal class could only be done by coming out to the front. I have used it for teaching on a much smaller scale -- tutorial sessions with 1 or 2 others, where everyone can be in Floor mode and can make their own spontaneous changes to what's ging on without any bureaucracy. This can include having everyone participate in editing a file of R commands, so that everyone can see what the effect of doing this or that is. I have found it to work well, and to be appreciated by students. This mode could work with more than 1 or 2 students. The main snag with using XMX is that the X display must be the same on all machines. This was a deliberate design implementation, since it cuts down enormously on network traffic. However, it does mean that all machines must have a sufficiently large overlap in X resouces (i.e. as reported by 'xdpyinfo'). The resources don't have to be absolutely identical, but certainly colour-depths must be the same all round, and in my experience you had better have the same display geometry (e.g. 1280x1024), and there are various other things. One way to find out is to start it up nominating all the participating machines at once. If there's a compatible sufficient overlap, it will work. However, in a teaching Lab with all the machines the same, it should be straightforward to get it going. One nice thing about doing things this way is that a student could be running a private R session alongside the public XMX one, and copypaste from the public one into the private one (or vice versa). They could also be running an editor session to write their own course notes. Anyway, there it is if you hadn't heard of it. Best wishes to all, Ted. E-Mail: (Ted Harding) [EMAIL PROTECTED] Fax-to-email: +44 (0)870 094 0861 Date: 19-Sep-05 Time: 18:36:32 -- 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
Re: [R] factor as seq() in for loop
Tord Snall [EMAIL PROTECTED] writes: Dear all, I would like to use the values in vegaggr.BLMCMR02$colony str(vegaggr.BLMCMR02) `data.frame': 1678 obs. of 3 variables: $ vegtype : Factor w/ 27 levels 2010,2020,..: 3 4 5 19 4 5 19 5 $ colony: Factor w/ 406 levels 0,1,10,100,..: 1 1 1 1 2 2 2 $ Totvegproparea: num 0.00055 0.03956 0.95705 0.00284 0.05215 ... as the seq in a for loop (for(var in seq)), and therefore first picked them out as cols = aggregate(list(A = vegaggr.BLMCMR02$Totvegproparea), list(colony = vegaggr.BLMCMR02$colony), sum)[,1] str(cols) Factor w/ 406 levels 0,1,10,100,..: 1 2 3 4 5 6 7 8 9 10 ... I next planned to transform cols using as.integer(cols). However, this transformation gives a vector corresponding to seq(from=1,to=length(cols)). Yes, it would do that... The construction looks a bit complicated, though. If you want the levels of the factor, wouldn't levels(vegaggr.BLMCMR02$colony) do it? Could someone please give advice on how to make use of a factor as the seq in a for loop (which is strictly not allowed' according to ?Control). You're not exactly being clear about what the for loop should achieve! -- 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] factor as seq() in for loop
On 9/19/05, Tord Snall [EMAIL PROTECTED] wrote: Dear all, I would like to use the values in vegaggr.BLMCMR02$colony str(vegaggr.BLMCMR02) `data.frame': 1678 obs. of 3 variables: $ vegtype : Factor w/ 27 levels 2010,2020,..: 3 4 5 19 4 5 19 5 $ colony: Factor w/ 406 levels 0,1,10,100,..: 1 1 1 1 2 2 2 $ Totvegproparea: num 0.00055 0.03956 0.95705 0.00284 0.05215 ... as the seq in a for loop (for(var in seq)), and therefore first picked them out as cols = aggregate(list(A = vegaggr.BLMCMR02$Totvegproparea), list(colony = vegaggr.BLMCMR02$colony), sum)[,1] str(cols) Factor w/ 406 levels 0,1,10,100,..: 1 2 3 4 5 6 7 8 9 10 ... I next planned to transform cols using as.integer(cols). However, this transformation gives a vector corresponding to seq(from=1,to=length(cols)). Could someone please give advice on how to make use of a factor as the seq in a for loop (which is strictly not allowed' according to ?Control). What are you trying to loop over? Species - head(iris$Species) for(i in seq(Species)) print(Species[i]) will loop over the elements of the factor Species and you can use for(lev in levels(Species)) print(lev) to loop over its levels. __ 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] library (tree)- which samples belong to each terminal branch?
Dear all What I would like to know is if there is a way to know which specific samples fall within a terminal branch created by tree? The classification summarizes data, but I want to know which specific samples fall within each classification (branch in the dendrogram). Data contains site codes and multiple fields describing the characters of a single specimen of grass growing there. Each sample is an individual plant from a site, all plants are similar to each other, but may represent different species. I classified species using tree and these are labeled by site name. The classification gives me a nice separation of supposed species in terms of only 3 sites but with some sites occurring on more than one terminal branch, this obviously summarizes one or more species/sites on each branch of the dendrogram. So back to my question, how do I find out which samples fall within each branch? Cheers Chris Buddenhagen, Botany Department, Charles Darwin Research Station, Santa Cruz,Galapagos. Mail: Charles Darwin Foundation, Casilla 17-01-3891 Avenida 6 de Diciembre N36-109 y Pasaje California Quito, ECUADOR __ EL CONTENIDO DE ESTE MENSAJE ES DE ABSOLUTA RESPONSABILIDAD DEL AUTOR. FUNDACION CHARLES DARWIN WWW.DARWINFOUNDATION.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
[R] waveform filtering
I'm not an engineer so I hope I'm using the correct terminology here. I have a recorded waveform that I want to apply low and high pass filters too, are tehre already R functions existing to do this or am I going to have to program my own? thanks for any pointers tom __ 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] how to extract the column name or value from the numerical value of the matrix
Dear sir, i have a matrix like x-c(1:20) A-matrix(x,4,5) A [,1] [,2] [,3] [,4] [,5] [1,]159 13 17 [2,]26 10 14 18 [3,]37 11 15 19 [4,]48 12 16 20 I want to extract the column value for the matrix value 11... or the row value for 14.. how it is possible? thanks for your help with regards, boopathy. Thirumalai Shanmuha Boopathy, Zimmer no : 07-15, Rütscher strasse 165, 52072 Aachen . Germany. Home zone : 0049 - 241 - 9813409 Mobile zone : 0049 - 176 - 23567867 __ 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] library (tree)- which samples belong to each terminal branch?
On Mon, 19 Sep 2005, Chris Buddenhagen wrote: What I would like to know is if there is a way to know which specific samples fall within a terminal branch created by tree? The classification summarizes data, but I want to know which specific samples fall within each classification (branch in the dendrogram). Please read the help page (as we do ask). Component `where' in the returned object tells you the information you are seeking. If you want this for new data, predict.tree has a type=where to tell you. Data contains site codes and multiple fields describing the characters of a single specimen of grass growing there. Each sample is an individual plant from a site, all plants are similar to each other, but may represent different species. I classified species using tree and these are labeled by site name. The classification gives me a nice separation of supposed species in terms of only 3 sites but with some sites occurring on more than one terminal branch, this obviously summarizes one or more species/sites on each branch of the dendrogram. So back to my question, how do I find out which samples fall within each branch? -- 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
Re: [R] how to extract the column name or value from the numerical value of the matrix
look at ?col and ?row. One way to use them is: col(A)[A==11] row(A)[A==14] hope this helps, Greg Snow, Ph.D. Statistical Data Center, LDS Hospital Intermountain Health Care [EMAIL PROTECTED] (801) 408-8111 shanmuha boopathy [EMAIL PROTECTED] 09/19/05 01:30PM Dear sir, i have a matrix like x-c(1:20) A-matrix(x,4,5) A [,1] [,2] [,3] [,4] [,5] [1,]159 13 17 [2,]26 10 14 18 [3,]37 11 15 19 [4,]48 12 16 20 I want to extract the column value for the matrix value 11... or the row value for 14.. how it is possible? thanks for your help with regards, boopathy. Thirumalai Shanmuha Boopathy, Zimmer no : 07-15, Rütscher strasse 165, 52072 Aachen . Germany. Home zone : 0049 - 241 - 9813409 Mobile zone : 0049 - 176 - 23567867 __ 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
Re: [R] how to extract the column name or value from the numerical value of the matrix
On Mon, 2005-09-19 at 12:30 -0700, shanmuha boopathy wrote: Dear sir, i have a matrix like x-c(1:20) A-matrix(x,4,5) A [,1] [,2] [,3] [,4] [,5] [1,]159 13 17 [2,]26 10 14 18 [3,]37 11 15 19 [4,]48 12 16 20 I want to extract the column value for the matrix value 11... or the row value for 14.. how it is possible? thanks for your help with regards, boopathy. Why did you copy r-devel? r-help is the appropriate forum. See ?which and take note of the 'arr.ind' argument: which(A == 11, arr.ind = TRUE) row col [1,] 3 3 which(A == 14, arr.ind = TRUE) row col [1,] 2 4 HTH, Marc Schwartz __ 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] restore file magic number
Hello, I recently lost access to my workspace developed over a long time. When I try to load this work space (called Casmalia.Rdata) or double click on it to open it in R, I see a popup which says (Embedded image moved to file: pic17720.jpg) : And under this, the last line in the console commands is: Bad restore file magic number. (File may be corrupted) - no data loaded. The Casmalia.Rdata file looks the right size. But is apparently missing a magic number. Can this file be fixed in an editor such as Textpad so that it can be restored? --- Kristin Cothern, URS Corporation 130 Robin Hill Road Santa Barbara, California 93117 tel. 805-964-6010 This e-mail and any attachments are confidential. If you receive this message in error or are not the intended recipient, you should not retain, distribute, disclose or use any of this information and you should destroy the e-mail and any attachments or copies. __ 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] Change in behavior of compare statement
On 9/19/2005 1:40 PM, Duncan Murdoch wrote: On 9/19/2005 12:25 PM, Tomas Andersson wrote: Dear all, I have come across some Windows Script code which calls the file.exists function in R to check for the existence of a particular file on an R server. This is what the code looks like: do { // some useful code } while (m_workspace.session.eval(file.exists( + dataFile + )) != 1); What is this function m_workspace.session.eval? It's not a base function. Perhaps its definition has changed? This has to be one of the most out-to-lunch answers I ever wrote. Sorry, please ignore it. The code above is Windows Script, whatever that is, not R. file.exists hasn't changed, but whatever interface you're using to R presumably has. In the code you're looking at there, it converted TRUE to 1, and now it doesn't. Duncan Murdoch It appears that the behavior of this code is different when used with R version 2.1.1 compared to R version 1.9.1. If the compare statement s is changed from while (s != 1) to while (s != True) or while !(s) the behavior of the code is the same with both versions of R. After reading the R documentation, I have not found any evidence that the original syntax (s != 1) is in any way recommended or supported (in version 2.1.1). Still, my questions are: s != 1 is perfectly valid R code. It will evaluate to TRUE in at least the following case: as.character(s) is not an NA, but isn't 1. s != True will evaluate to TRUE in most cases, unless s happens to contain that exact string. 1. Is it possible that the syntax has been supported in older versions of R? 2. Is the reason for the recent change in behavior known and documented anywhere? The syntax is fine, but it looks like a behaviour change in that function m_workspace.session.eval. You'll need to ask the author of that function what is happening. Duncan Murdoch __ 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] restore file magic number
On 9/19/2005 4:29 PM, [EMAIL PROTECTED] wrote: Hello, I recently lost access to my workspace developed over a long time. When I try to load this work space (called Casmalia.Rdata) or double click on it to open it in R, I see a popup which says (Embedded image moved to file: pic17720.jpg) : And under this, the last line in the console commands is: Bad restore file magic number. (File may be corrupted) - no data loaded. The Casmalia.Rdata file looks the right size. But is apparently missing a magic number. Can this file be fixed in an editor such as Textpad so that it can be restored? It's probably missing a lot more than a magic number. That's just a signal that the file format isn't really an R workspace any more. You need to get an old copy from a backup. If you're desperate and short of a backup, you could try looking at the file in a binary dump program to see if there's anything left of the contents (perhaps only part of it got trashed), and maybe try to restore just the good stuff, but it's not going to be easy, and the only real documentation for what to do is to look at the R source code. Duncan Murdoch __ 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] text editor TinR?
Greetings, Please help me remember the name of the tiny text editor that works with R TinR maybe? I cannot find it at all, and cannot remember it, it is really frustrating... Thanks, Mihai Nica, ABD Jackson State University 170 East Griffith Street G5 Jackson, MS 39201 601 914 0361 __ 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] pairwise comparisons among treatments
Hello R listing, I did two-way anova on lm. Further question the investigator interested in is what two treatments are different? I am looking for a command which could do pairwise comparison for every treatment. Could anyone help me out? Thanks a bunch Kevin __ 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] text editor TinR?
Dear Mihai, It's Tinn-R (with two n's), at http://www.sciviews.org/Tinn-R/. 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 Mihai Nica Sent: Monday, September 19, 2005 4:07 PM To: r-help@stat.math.ethz.ch Subject: Re: [R] text editor TinR? Greetings, Please help me remember the name of the tiny text editor that works with R TinR maybe? I cannot find it at all, and cannot remember it, it is really frustrating... Thanks, Mihai Nica, ABD Jackson State University 170 East Griffith Street G5 Jackson, MS 39201 601 914 0361 __ 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] text editor TinR?
Mihai Nica [EMAIL PROTECTED] writes: Greetings, Please help me remember the name of the tiny text editor that works with R TinR maybe? I cannot find it at all, and cannot remember it, it is really frustrating... http://www.sciviews.org/Tinn-R/ -- 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] waveform filtering
You should probably have a look at the sound packages for R, tuneR and sound, I believe, on http://cran.r-project.org. Applying a filter can be done with filter(), but you need to come up with filter coefficients. High-pass and low-pass have simple descriptions in the Fourier transform space, so you might want to specify the frequency response of your filter directly there, then do an inverse Fourier transform (fft() in R) to get coefficients. The ingredients are all there in R itself; but the packages tuneR and sound might have exactly what you want. A book on time series or signal processing might be helpful. Reid Huntsinger -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of tom wright Sent: Monday, September 19, 2005 10:36 AM To: R-help mailing list Subject: [R] waveform filtering I'm not an engineer so I hope I'm using the correct terminology here. I have a recorded waveform that I want to apply low and high pass filters too, are tehre already R functions existing to do this or am I going to have to program my own? thanks for any pointers tom __ 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
Re: [R] pairwise comparisons among treatments
help.search('pairwise comparison') -- Bert Gunter Genentech Non-Clinical Statistics South San Francisco, CA The business of the statistician is to catalyze the scientific learning process. - George E. P. Box -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of kevin Lin Sent: Monday, September 19, 2005 2:33 PM To: r-help@stat.math.ethz.ch Subject: [R] pairwise comparisons among treatments Hello R listing, I did two-way anova on lm. Further question the investigator interested in is what two treatments are different? I am looking for a command which could do pairwise comparison for every treatment. Could anyone help me out? Thanks a bunch Kevin __ 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
Re: [R] pairwise comparisons among treatments
Hello Bert, Thanks for your response. I don't know why I am getting following error messages. I use version of R version 2.1.1, 2005-06-20, powerpc-apple-darwin7.9.0 in Mac OS X version 10.3.8. I did try to load following packages before, but it failed. help.search(pairwise comparison) Error in help.search(pairwise comparison) : could not find package 'arrayQuality.1' In addition: Warning messages: 1: no Rd contents for package 'R2HTML' in '/Library/Frameworks/R.framework/Resources/library' in: help.search(pairwise comparison) 2: no Rd contents for package 'affylmGUI' in '/Library/Frameworks/R.framework/Resources/library' in: help.search(pairwise comparison) Thanks. Kevin help.search('pairwise comparison') -- Bert Gunter Genentech Non-Clinical Statistics South San Francisco, CA The business of the statistician is to catalyze the scientific learning process. - George E. P. Box __ 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] text editor TinR?
Thanks to everybody answering my call for help. It is Tinn-R! Mihai Nica, ABD Jackson State University 170 East Griffith Street G5 Jackson, MS 39201 601 914 0361 __ 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] How to mimic pdMat of lme under lmer?
On 9/19/05, Joris De Wolf [EMAIL PROTECTED] wrote: Dear members, I would like to switch from nlme to lme4 and try to translate some of my models that worked fine with lme. I have problems with the pdMat classes. Below a toy dataset with a fixed effect F and a random effect R. I gave also 2 similar lme models. The one containing pdLogChol (lme1) is easy to translate (as it is an explicit notation of the default model) The more parsimonious model with pdDiag replacing pdLogChol I cannot reproduce with lmer. The obvious choice for me would be my model lmer2, but this is yielding different result. Somebody any idea? Thanks, Joris I am using R version 2.1.0 for Linux and the most recent downloads of Matrix and nlme #dataset from McLean, Sanders and Stroup F - factor(c(1,1,1,1,1,1,2,2,2,2,2,2)) R - factor(c(1,1,2,2,3,3,1,1,2,2,3,3)) s - c(51.43,51.28,50.93,50.75,50.47,50.83,51.91,52.43,52.26,52.33,51.58,51.23) DS - data.frame(F,R,s) DS$F - as.factor(DS$F) DS$R - as.factor(DS$R) library(nlme) lme1 - lme(data = DS,s ~ F,random = list(R = pdLogChol(~F))) lme2 - lme(data = DS,s ~ F,random = list(R = pdDiag(~F))) summary(lme1) summary(lme2) library(lme4) lmer1 - lmer(data = DS,s ~ F + (F|R)) lmer2 - lmer(data = DS,s ~ F + (1|R) + (1|F)) summary(lmer1) summary(lmer2) You need to generate the two columns of the indicator matrix for F as separate model matrices. This will involve some awkward construction like (fm1 - lmer(s ~ F +(F|R), DS)) Linear mixed-effects model fit by REML Formula: s ~ F + (F | R) Data: DS AIC BIClogLik MLdeviance REMLdeviance 20.69259 23.60203 -4.3462976.03915 8.692593 Random effects: Groups NameVariance Std.Dev. Corr R(Intercept) 0.108796 0.32984 F2 0.102008 0.31939 -0.014 Residual 0.048525 0.22028 # of obs: 12, groups: R, 3 Fixed effects: Estimate Std. Error DF t value Pr(|t|) (Intercept) 50.9483 0.2106 10 241.9188 2.2e-16 F21.0083 0.2240 10 4.5014 0.001141 (fm2 - lmer(s ~ F + (0+as.numeric(F==1)|R) + (0+as.numeric(F==2)|R), DS)) Linear mixed-effects model fit by REML Formula: s ~ F + (0 + as.numeric(F == 1) | R) + (0 + as.numeric(F == 2) | R) Data: DS AIC BIClogLik MLdeviance REMLdeviance 19.62621 22.05075 -4.813107 7.439584 9.626215 Random effects: Groups Name Variance Std.Dev. Ras.numeric(F == 1) 0.108796 0.32984 Ras.numeric(F == 2) 0.207896 0.45596 Residual0.048525 0.22028 # of obs: 12, groups: R, 3; R, 3 Fixed effects: Estimate Std. Error DF t value Pr(|t|) (Intercept) 50.948330.21060 10 241.9188 2e-16 F2 1.008330.34891 10 2.8899 0.01611 There are probably more elegant ways of doing this but I don't think there is any really clean way. If you want to assume that all the variances are equal then you can estimate the model using an interaction. (fm3 - lmer(s ~ F + (1|F:R), DS)) Linear mixed-effects model fit by REML Formula: s ~ F + (1 | F:R) Data: DS AIC BIClogLik MLdeviance REMLdeviance 17.77917 19.7188 -4.889587 7.669022 9.779175 Random effects: Groups NameVariance Std.Dev. F:R (Intercept) 0.158346 0.39793 Residual 0.048525 0.22028 # of obs: 12, groups: F:R, 6 Fixed effects: Estimate Std. Error DF t value Pr(|t|) (Intercept) 50.948330.24672 10 206.5049 2e-16 F2 1.008330.34891 10 2.8899 0.01611 confidentiality notice: The information contained in this e-mail is confidential and...{{dropped}} __ 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
Re: [R] Teaching R - In front of the computer?
From the prospective of a student, I highly recommend the hands on approach. I had the best outcomes when I was provided with examples, and then encouraged to make my own modifications in a lab / class environment, with help available. This approach allows students to build the skills required to program independently, without becoming overwhelmed at the start. Murray Murray Pung | Research Analyst AIM Research HR Consulting PO Box 328, Nth Sydney NSW 2060 P +61 (02) 9956 3951 F +61 (02) 9922 2210 www.aimsurveys.com.au -Original Message- From: John Fox [mailto:[EMAIL PROTECTED] Sent: Tuesday, 20 September 2005 4:25 AM To: Rau, Roland Cc: r-help@stat.math.ethz.ch Subject: Re: [R] Teaching R - In front of the computer? Dear Roland, I've taught the use of R to this kind of audience many times. Take a look at http://socserv.socsci.mcmaster.ca/jfox/Courses/UCLA/index.html for materials used in such a workshop, and at http://socserv.socsci.mcmaster.ca/jfox/Teaching-with-R.pdf for a paper on teaching social statistics with R. As others have suggested, using static slides is not a good idea, and having at least a live display for the presenter is essential. It also helps to have the students sitting at computers and able to try things out for themselves. If this is a workshop devoted to R, I'd strongly recommend this format. On the other hand, if you're teaching R in the context of a more general statistics course, you can cover the basics in a hands-on workshop and then use the LCD projector to introduce new commands, etc., during the course as they're needed. I find that once they've acquired the basics, students are able to work more independently. I hope this helps, John On Mon, 19 Sep 2005 15:25:14 +0200 Rau, Roland [EMAIL PROTECTED] wrote: Dear R-Users, given you have been teaching R to students (grad level, mainly social science background, no previous programming experience, 80% know SPSS), what are your experiences concerning the style of teaching? Do you prefer to stand in front of the class like in normal lectures and you show them slides? Or do you you explain some concept (for example things like mydata[order(var1, var2),]) and show it directly on the computer via beamer/projector and also the students have to enter it on the computers in front of them. Any experiences you can share are highly appreciated. Thanks, Roland + This mail has been sent through the MPI for Demographic Rese...{{dropped}} __ 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 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 __ 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] Are least-squares means useful or appropriate?
Estimado Felipe: If you provide a very simple example (as suggested in the posting guide, www.R-project.org/posting-guide.html), it would allow those of use who rarely use SAS to respond. Try to think of the simplest possible toy data set and analysis that shows the difference between the SAS answer and the answer you get from a certain R function. If you post something simple of that nature that someone can copy from your email into R and try other things in a minute or two, it will likely increase the speed and utility of a reply. Buena Suerte, spencer graves Felipe wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi. I have been reading about the convenience of using least-squares means (a. k. a. adjusted means) in multiple comparisons (I used to resort to them when using SAS). I even read a post in this list warning against them, but not giving much detail. What do you think about this? Greetings. Felipe -BEGIN PGP SIGNATURE- iEYEARECAAYFAkMqq5gACgkQWtdQtNzjBl4AigCfQJ64O0wrdYK/1iMReW5RtI1d tMIAn3DQSdk+4D7AK7VQGtWo0TElrFG7 =j9EX -END PGP SIGNATURE- __ 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] logistic regression with nominal predictors
This sounds to me like a great research project that could be answered relatively easily with a Monte Carlo study. An execllent mathematician might be able to produce simple theoretical limits on the error from using ranks or normal scores, but such limits would likely be much wider than one could get in a typical case using Monte Carlo. And Monte Carlo in R is normally fairly easy even for quite complicated situations. spencer graves Ramón Casero Cañas wrote: (Sorry for obvious mistakes, as I am quite a newby with no Statistics background). My question is going to be what is the gain of logistic regression over odds ratios when none of the input variables is continuous. My experiment: Outcome: ordinal scale, ``quality'' (QUA=1,2,3) Predictors: ``segment'' (SEG) and ``stress'' (STR). SEG is nominal scale with 24 levels, and STR is dychotomous (0,1). Considering the outcome continuous, two-way ANOVA with aov(as.integer(QUA) ~ SEG * STR) doesn't find evidence of interaction between SEG and STR, and they are significant on their own. This is the result that we would expect from clinical knowledge. I use xtabs(~QUA+SEG, data=data2.df, subset=STR==0) xtabs(~QUA+SEG, data=data2.df, subset=STR==0) for the contingency tables. There are zero cells, and for some values of SEG, there is only one none-zero cell, i.e. some values of SEG determine the output with certainty. So initially I was thinking of a proportional odds logistic regression model, but following Hosmer and Lemeshow [1], zero cells are problematic. So I take out of the data table the deterministic values of SEG, and I pool QUA=2 and QUA=3, and now I have a dychotomous outcome (QUA = Good/Bad) and no zero cells. The following model doesn't find evidence of interaction glm(QUA ~ STR * SEG, data=data3.df, family=binomial) so I go for glm(QUA ~ STR + SEG, data=data3.df, family=binomial) (I suppose that what glm does is to create design variables for SEG, where 0 0 ... 0 is for the first value of SEG, 1 0 ... 0 for the second value, 0 1 0 ... 0 for the third, etc). Coefficients: Estimate Std. Error z value Pr(|z|) (Intercept) -1.085e+00 1.933e-01-5.614 1.98e-08 *** STR.L2.112e-01 6.373e-02 3.314 0.000921 *** SEGP2C.MI -9.869e-01 3.286e-01-3.004 0.002669 ** SEGP2C.AI -1.306e+00 3.585e-01-3.644 0.000269 *** SEGP2C.AA -1.743e+00 4.123e-01-4.227 2.37e-05 *** [shortened] SEGP4C.ML -5.657e-01 2.990e-01-1.892 0.058485 . SEGP4C.BL -2.908e-16 2.734e-01 -1.06e-15 1.00 SEGSAX.MS1.092e-01 2.700e-01 0.405 0.685772 SEGSAX.MAS -5.441e-16 2.734e-01 -1.99e-15 1.00 SEGSAX.MA7.130e-01 2.582e-01 2.761 0.005758 ** SEGSAX.ML1.199e+00 2.565e-01 4.674 2.96e-06 *** SEGSAX.MP1.313e+00 2.570e-01 5.108 3.26e-07 *** SEGSAX.MI8.865e-01 2.569e-01 3.451 0.000558 *** --- Signif. codes: 0 `***' 0.001 `**' 0.01 `*' 0.05 `.' 0.1 ` ' 1 (Dispersion parameter for binomial family taken to be 1) Null deviance: 3462.0 on 3123 degrees of freedom Residual deviance: 3012.6 on 3101 degrees of freedom AIC: 3058.6 Number of Fisher Scoring iterations: 6 Even though some coefficients have no evidence of statistical significance, the model requires them from a clinical point of view. At this point, the question would be how to interpret these results, and what advantage they offer over odds ratios. From [1] I can understand that in the case of a dychotomous and a continuous predictor, you can adjust for the continuous variable. But when all predictors are dychotomous (due to the design variables), I don't quite see the effect of adjustment. Wouldn't it be better just to split the data in two groups (STR=0 and STR=1), and instead of using logistic regression, use odds ratios for each value of SEG? Cheers, Ramón. [1] D.W. Hosmer and S. Lemeshow. ``Applied Logistic Regression''. John-Wiley. 2000. -- 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] Transform variable number of rows per subject to column variables?
Hello, I am very new to R, but I am having trouble with my dataset. I have a data frame where a subject has a variable number of multiple observations for each row, which I wish the transform these observations to column variables. An example of the data frame ID TEST.A TEST.B 1 10 1 1 13 2 1 11 1 2 15 2 2 17 3 And I wish to transform it to the following: ID TEST.A1 TEST.A2 TEST.A3 TEST.B1 TEST.B2 TEST.B3 1 10 13 11 1 2 1 2 15 17 NA 2 3 NA In other words, for the variable number of repeated follow up studies, a new column variable for each subject, but they are grouped by the original test. Thank you for any help - I'm realizing that I am a terrible programmer! Bing Ho __ 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] nls()
I have passed the identities and values of fixed parameters via the ... arguments in optim; I don't know about nls. Then internal to the function that optim is to minimize, I combine the x argument with the fixed parameters to obtain the full set of parameters. I've used that effectively. It's not trivial, but it can be made to work. spencer graves J.M. Breiwick wrote: Hi, I am using nls() with the form: nls(~my.fcn(...)) because I have to iteratively compute the expected y values. The function my.fcn() returns y.obs-y.pred However, I want to fix some of the parameters in my.fcn at various values and compute the parameter estimates. In Splus there is such a thing as a parameterized dataframe. I don't think this exists in R so does anyone know how to set one or more of the parameters as constants in the model? Thank you. Jeff Breiwick __ 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] Transform variable number of rows per subject to column variables?
On 9/20/05, Bing Ho [EMAIL PROTECTED] wrote: Hello, I am very new to R, but I am having trouble with my dataset. I have a data frame where a subject has a variable number of multiple observations for each row, which I wish the transform these observations to column variables. An example of the data frame ID TEST.A TEST.B 1 10 1 1 13 2 1 11 1 2 15 2 2 17 3 And I wish to transform it to the following: ID TEST.A1 TEST.A2 TEST.A3 TEST.B1 TEST.B2 TEST.B3 1 10 13 11 1 2 1 2 15 17 NA 2 3 NA In other words, for the variable number of repeated follow up studies, a new column variable for each subject, but they are grouped by the original test. First manufacture a time column and then use reshape: tt - sequence(rle(DF$ID)$lengths) reshape(cbind(tt, DF), idvar = ID, timevar = tt, direction = wide) Another possibility is to use the reshape package: library(reshape) DF.d - deshape(cbind(tt, DF), id = 1:2) # same tt as above reshape(DF.d, ID ~ variable + tt) __ 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