similar to: question about scalability of R

You could break your 3 class problem into several (2 or 3) 2 class problems, and then use Andy's suggestion (see the CART book). There are several ways to break the problem into 2 class problems, and several ways to combine the resulting classifiers. Tom Dietterich, Jerry Friedman, Trevor Hastie and Rob Tibshirani, among others, have articles on the question, in places like Annals of

Hi, there: I am wondering if I can find some detailed explanation on gbm or explanation on examples of gbm. thanks, Ed

Hi, there: I am looking for a regression modeling (like regression trees) approach for a large-scale industry dataset. Any suggestion on a package from R or from other sources which has a decent accuracy and scalability? Any recommendation from experience is highly appreciated. Thanks, Weiwei -- Weiwei Shi, Ph.D "Did you always know?" "No, I did not. But I believed..."

Hi there, I need a network builder and it can change the node size and color; I am not sure if network package in R can do this or not. The other functions I wanted have been found in that package. BTW, if there is another package in R relating to this, please suggest too. Thanks, Weiwei -- Weiwei Shi, Ph.D Research Scientist "Did you always know?" "No, I did not. But I

Hi, there: I made a function to do k-fold cross-validation as below. Basically whenever I call cv(test) for example, an error message like: 20Fold 1 Error in model.frame(formula, rownames, variables, varnames, extras, extranames, : variable lengths differ please help. My test dataset has 142 variables, the last one is a categorical response variable. also, i am not sure how to save

Hi, there: Suppose I have a couple of data.frames and each one has five columns (one for x-axis, two for y-axis and two for std of y's.) There is another dimensions (besides x and y) which is continuous. My question is, how to plot such series of data frames in one plot (thus, 3-dimensional plot) AND multiple 2-D plots. I am not familar with R's plotting utilities. Thanks. -- Weiwei

FWIW, I wrote a little function to extract variable importance as defined in the CART book a while ago. It's rather limited: Only works for regression problem, and you need to set maxsurrogate=0 and maxcompete=0. It may (or may not) help you: varimp.rpart <- function(x) { dev <- x$frame[, c("var", "dev")] dev <- dev[dev$var != "<leaf>",

Excuses I have a mistake in previous mail Type stats:::t.test.defaultThe formal way is to use getAnywhere(t.test) Justin BEM Elève Ingénieur Statisticien Economiste BP 294 Yaoundé. Tél (00237)9597295. ----- Message d'origine ---- De : justin bem <justin_bem@yahoo.fr> À : Weiwei Shi <helprhelp@gmail.com> Cc : R-help@stat.math.ethz.ch Envoyé le : Mardi, 12 Décembre 2006,

Hi, I searched the archives and did not find a good solution to that. assume I have 10 sets and I want to have the common character elements of them. how could i do that? -- Weiwei Shi, Ph.D Research Scientist GeneGO, Inc. "Did you always know?" "No, I did not. But I believed..." ---Matrix III

Hi, there: I am wondering if there is a quick way to "reverse" a list like this: t0 <- list(a=1, b=1, c=2, d=1) reverst t0 to t1 > t1 $`1` [1] "a" "b" "d" $`2` [1] "c" thanks. -- Weiwei Shi, Ph.D Research Scientist GeneGO, Inc. "Did you always know?" "No, I did not. But I believed..." ---Matrix III

Hi, there: Assume I have a dataframe with rownames like A with rownames like a to e, > A [,1] [,2] a 1 6 b 2 7 c 3 8 d 4 9 e 5 10 when I use A[1,], I lost the rowname for it, like below. How could I keep it? Is there an easy way instead that I have to modify by myself after I used A[1,] manually. > A[1,] [1] 1 6 Thanks, W. -- Weiwei Shi, Ph.D Research

hi, if I have 20 x 3 data.frame, how to split it into 10 x 6 (moving the lower part of 10x3 to column) or 5 x 12 thanks -- Weiwei Shi, Ph.D Research Scientist GeneGO, Inc. "Did you always know?" "No, I did not. But I believed..." ---Matrix III

hi, is there some good summary on clustering methods in R? It seems there are many packages involving it. And I have two questions on clustering here: 1. Is there a way of evaluate the effecitives (or seperation) of clustering (rather than by visualization)? 2. Is there a search method (like genetic search) which can help find the best subset of attributes which gives best seperation? Thanks,

Hi, I am wondering how to make a function Fun to make the following work: t0 <- (paste("hgu133a", "ENTREZID", sep="")) xx <- as.list(Fun(t0)) # make it work like xx<-as.list(hgu133aENTREZID) thanks, -- Weiwei Shi, Ph.D Research Scientist GeneGO, Inc. "Did you always know?" "No, I did not. But I believed..." ---Matrix III

Hi, I have a question on read.table. I have a dataset with 273,000 lines and 195 columns. I used the read.table to load the data into R: trn<-read.table('train1.dat', header=F, sep='|', na.strings='.') I found it takes forever. then I run 1/10 of the data (test) using read.table again. And this time it finished quickly. So, there might be something wrong in my data

Hi, there: I spent some time on this but I think I really cannot figure it out, maybe I missed something here: my data looks like this: > dim(trn3) [1] 7361 209 > dim(val3) [1] 7427 209 > mg.rf2<-randomForest(x=trn3[,1:208], y=trn3[,209], data=trn3, xtest=val3[, 1:208], ytest=val3[,209], importance=T) my test data has 7427 observations but after prediction, > dim(mg.rf2$votes)

Hi, I have a dataset and want to build a generalized linear model on it. Unfortunately, complete.cases(df) returns null, which means I have to find a way to "fill" those missings. One way is following my previous post to use median to replace(or use most freq. of level to replace for catergorical case), but I am wondering if there are other ways, when glm or something like it is

Dear listers, a little off-topic: I am looking for and compare algorithms which can calculate "distance" or "similarity" between two gene lists with different lengths. Any paper, any implementation in R and any suggestion is welcome! Thanks, -- Weiwei Shi, Ph.D Research Scientist GeneGO, Inc. "Did you always know?" "No, I did not. But I believed..."

Stock returns and other financial data have often found to be heavy-tailed. Even Cauchy distributions (without even a first absolute moment) have been entertained as models. Your qq function subtracts numbers on the scale of a normal (0,1) distribution from the input data. When the input data are scaled so that they are insignificant compared to 1, say, then you get essentially the

Hi, forgive me if it is due to my "laziness" :) I am wondering if there are functionalities in R, which can do like "more" and "tab" in linux: more(one.data.frame) so I can browse through it. Sometimes I can use one.data.frame[1:100,], but still not as good as "more" in linux. tab: can I use tab to auto complete an defined object name in R so I don't