[R] Mclust help
Hi, quick question from a social scientist new to R: how do I add labels/row names to the nodes in coordProj / plot.Mclust plots? Apologies if this is a silly question... Thanks. -- Bobby Sick sense of humor? Visit Yahoo! TV's Comedy with an Edge to see what's on, when. __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] 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
[R] random effects plot
Hi, does anyone know how to do random effects plots for multi-level models? plot(ranef(model)) doesn't seem to work in the latest release. Here's the error message: Error in table(x) : argument x is missing, with no default In addition: Warning message: The 'formula' argument has been renamed to 'x'. See ?xyplot in: dotplot(formula = .groups ~ .pars | .enames, data = list(.pars = c(-0.0388163041843026, Help a poor sociologist out... -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] survey weights
Thanks, Thomas. Yes, that's exactly what happened: the warnings came first after data(China), and then after dchina-svydesign... So the design object isn't being produced? The dataset is very large, and the weights were already set in Stata before importing. Would either of those cause problems? -Bobby --- Thomas Lumley [EMAIL PROTECTED] wrote: On Sat, 3 Sep 2005, A Das wrote: Hi all, I've been trying to get a large (12mb) Stata survey database into R. I managed that, but when I attach survey weights, something goes wrong. The error message is: object dchina not found. Here's the script: If that is the *first* message then something extremly strange is happening library(car) library(foreign) library(survey) China - read.dta(C:/final07c2.dta) attach(China) This attach() isn't necessary or helpful data(China) You should get a warning here Warning message: data set 'China' not found in: data(China) since China isn't one of the built-in data sets. If you don't get this message it suggests that you do have a built-in dataset called China, which will have overwritten your file. dchina-svydesign(id=~psu,strata=~strata,weights=~weight0x, data=China,nest=TRUE) If this line doesn't produce an error message then a variable called dchina must have been produced, in which case you shouldn't get an error message saying it wasn't found in the next line. summary(dchina) Are you sure there wasn't an earlier error message from the call to svydesign()? -thomas __ 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] survey weights
Just: missing values in object. That would imply the object was created. But then I write dchina, and it says object dchina not found. -Bobby --- Thomas Lumley [EMAIL PROTECTED] wrote: On Sun, 4 Sep 2005, A Das wrote: Thanks, Thomas. Yes, that's exactly what happened: the warnings came first after data(China), and then after dchina-svydesign... So the design object isn't being produced? The dataset is very large, and the weights were already set in Stata before importing. Would either of those cause problems? Probably not. What was the error message from svydesign()? That is what will say what went wrong. -thomas __ 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] survey weights
That worked. Many thanks, Thomas. -Bobby --- Thomas Lumley [EMAIL PROTECTED] wrote: On Sun, 4 Sep 2005, A Das wrote: Just: missing values in object. That would imply the object was created. But then I write dchina, and it says object dchina not found. No, it would not imply the object was created. If it was an error message (rather than a warning) the object would not have been created. I presume the full message was Error in na.fail.default(object) : missing values in object If so, it sounds as though you have missing values in the id, weights, or strata variable. summary(China[,c(psu,stata,weight0x]) will verify this. Stata will just have dropped these observations (use -svydes- to verify this). If you want to drop the observations in R you need to do this explicitly. Having missing data may be unavoidable, but if you have observations in a sample it seems that you should know how they were sampled. To drop these observations you could use obsChina - subset(China, !is.na(psu) !is.na(strata) !is.na(weight0x)) and then use obsChina rather than China in the svydesign() function. -thomas -Bobby --- Thomas Lumley [EMAIL PROTECTED] wrote: On Sun, 4 Sep 2005, A Das wrote: Thanks, Thomas. Yes, that's exactly what happened: the warnings came first after data(China), and then after dchina-svydesign... So the design object isn't being produced? The dataset is very large, and the weights were already set in Stata before importing. Would either of those cause problems? Probably not. What was the error message from svydesign()? That is what will say what went wrong. -thomas Start your day with Yahoo! - make it your home page http://www.yahoo.com/r/hs Thomas Lumley Assoc. Professor, Biostatistics [EMAIL PROTECTED] University of Washington, Seattle __ 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] survey weights
Hi all, I've been trying to get a large (12mb) Stata survey database into R. I managed that, but when I attach survey weights, something goes wrong. The error message is: object dchina not found. Here's the script: library(car) library(foreign) library(survey) China - read.dta(C:/final07c2.dta) attach(China) data(China) dchina-svydesign(id=~psu,strata=~strata,weights=~weight0x,data=China,nest=TRUE) summary(dchina) Any thoughts? -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
[R] decision values and probability in SVM
Hi, I am using SVM from e1071 package. I can get decision values very easily. But whenever, I try to get the probability measure, it returns NULL. I use the following codes to generate decision.values and probability. Is there anything wrong in it? predictor-svm(train[,c(x1, x2, x3)], train[,x4], probability=TRUE) pred-predict(predictor, test[,c(x1, x2, x3)], probability=TRUE, decision.values=TRUE) attr(pred, decision.values)[1,] attr(pred, probabilities)[1,] Thanks for your help. Raj __ 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] feature (attribute) selection
Hi, Look for dprep package for wrapper based feature selction that use lda, knn etc. Also you can use package rfe that implements recursive feature elimination using SVM. -Original Message- From: WeiWei Shi To: R-help@stat.math.ethz.ch Sent: 2/2/05 5:43 PM Subject: [R] feature (attribute) selection Hi, there: Recently, I read some papers about feature or attribute selection and most of them are discussed in the context of supervised learning. I knew Weka has implementation on some of them but I am wondering if there is any package available in R which can do this kind of job. Thanks for advice, Ed __ 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] probabilty calculation in SVM
Hi All, In package e1071 for SVM based classification, one can get a probability measure for each prediction. I like to know what is method that is used for calculating this probability. Is it calculated using logistic link function? Thanks for your help. Regards, Raj __ 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] using created R objects in a program
Hi, I am running a program in --vanilla mode. I would like to know how to load saved functions in the program that it calls. EVerytime I call those objects it returns error and complains that it cannot find the function object I am calling. Please let me know. Thanks in advance. Regards, Raj __ 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] predict.glm
Hi All, Sorry for this is a very naive question. I am trying to do binary classification (male vs female) using glm using following data: X1X2X3Class 2.34.52.1Male 0.93.2 1.6 Male 1.71.82.6Feamle I am trying to use predict.glm for prediction with type=respose which gives the predicted probabilities as per documentation. My question is: which of the two classes does this probability corresponds to? My understanding is that it is the probability of the class that each of the new data has. Is that correct? Thanks. Raj [[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] glm vs multinom
Dear Colleagues, I am doing two class classification using logistic regression. I realized that I can either use glm function or multinom function. I know multinom is used for multiclass classification. But if I was it for binary classification, I was wondering if there is an difference in the results compared to glm results. Thanks in advance. Regards, Raj [[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] classification using logistic regression
Dear Colleagues, I would like to do classification using logistic regression. Which R package can I use? Also is there any package for feature selection for logistic regression based method? Thanks a lot in advance. Happy holidays. Regards, Raj [[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] random numbers within a given range
Dear Colleagues, Is there a way to get random numbers within a given range? Regards and thanks in advance, Raj [[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] feature selection for SVM
Dear Colleagues, Is there any package in R/S-plus which does feature selection for SVM? I am working with 8 features. Since SVM is regularization based method, I am not sure if I need to do feature selection for it. In anycase, I would really like to know if there is any package for feature selection in SVM. Thanks. Raj [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] reading svm function in e1071
Hi, If I try to read the codes of functions in e1071 package, it gives me following error message. library(e1071) svm function (x, ...) UseMethod(svm) environment: namespace:e1071 predict.svm Error: Object predict.svm not found Can someone help me on this how to read the codes of the functions in the e1071 package? Thanks. Raj [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] SVM: quality of prediction
Hi, I am using SVM for two class classification purpose using RBF kernel. Once SVM builds a optimal model, it to used to predict classes for the new data. I was wondering if there is any metric or score that can be calculated from SVM which measures the quality of prediction for each new data point. So what I want to know is the can among different predictions, is there anyway to know which one is more accurate (according to classifier model) and which one is less accurate. Thanks. Raj [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] SVM distance
Hi, I am using SVM from e1071 package. I am using RBF kernel. I would like to know how I can get d, the perpendicular distance from a datapoint to the hyperplane, that SVM calculates in higher dimensional space to classify it. Although, this is not something that people usually use, but for my case I like to do something with that distance. Does anybody have any idea how to get it? Thanks. Raj __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Feature selection
Hi Edgar, Thanks. dprep module works fine. However, I have one question. How does SFFS or SFS function know the class definition in the data? Function call doesn't require to specify the column that identifies the class type (as shown inthe example). Can you eleborate on this? Thanks. Rajdeep - Original Message - From: Edgar Acuna [EMAIL PROTECTED] To: Rajdeep Das [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Sent: Monday, October 25, 2004 7:11 PM Subject: Re: [R] Feature selection Raj, look at academic.uprm.edu/eacuna/softw.htm, I built a library of R functions for data preprocessing tasks including feature selection for supervised classification. Please send me your comments. Edgar On Mon, 25 Oct 2004, Rajdeep Das wrote: Hello, I want to do feature selection for classification purpose (using lda). Can someone point me to any R package or S-plus package for this? Something like SFS or SFFS method would be useful for me. Thanks. Raj [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] test
[[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] test
[[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Feature selection
Hello, I want to do feature selection for classification purpose (using lda). Can someone point me to any R package or S-plus package for this? Something like SFS or SFFS method would be useful for me. Thanks. Raj [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html