similar to: R question

Hello, I was wondering if anyone knew of an R function/R code to plot bivariate (3 dimensional) partial dependence plots in random forests (randomForest package). It is apparently possible using the rgl package (http://esapubs.org/archive/ecol/E088/173/appendix-C.htm) or there may be a more direct function such as the pairplot() in MART (multiple additive regression trees)? Many

Is there a way to get the coordinates from a plot of an MCA object? [[alternative HTML version deleted]]

Hello, I'm trying to create a series of randomForest objects, basically in a loop like this: forests <- list(); for (level in 1:10) { # do some other things here # create a random forest forest <- randomForest( x = x.level, y = z.level, ntree = trees ); forests <- c(forests, forest); } But instead of creating a list of 10 forests, this creates a list

Hello R-User! I appologise in advance if this should also go into statistics but I am presently puzzled. I have a data.frame (about 300 rows and about 80 variables) and my variables are dichotomous factors, continuous (numerical) and ordered factors. I would like to calculate the linear correlation between every pair of my variables, because I would like to perform a logistic regression (glm())

Is there an Interaction function in the Random Forest algorithm in the current version of R? --------------------------------- [[alternative HTML version deleted]]

Dear community I'm pretty new to R and I'm trying to do a Point-baserial correlation for a nominal dichotomous variable with a interval scaled variable. It works fine, but the output just shows me the correlation and nothing else (p-Value would be important). I tried it with the following codes: - biseral.cor() - cor.biseral() - I also tried a polyserial() I've found on this

Hello R Gurus, I am perplexed by the different results I obtained when I ran code like this: set.seed(100) test1<-randomForest(BinaryY~., data=Xvars, trees=51, mtry=5, seed=200) predict(test1, newdata=cbind(NewBinaryY, NewXs), type="response") and this code: set.seed(100) test2<-randomForest(BinaryY~., data=Xvars, trees=51, mtry=5, seed=200, xtest=NewXs, ytest=NewBinarY) The

Hi, I had two questions regarding random forests for regression. 1) I have read the original paper by Breiman as well as a paper dicussing an application of random forests and it appears that the one of the nice features of this technique is good predictive ability. However I have some data with which I have generated a linear model using lm(). I can get an RMS error of 0.43 and an R^2 of

I am trying to run a GLMM on some binomial data. My fixed factors include 2 dichotomous variables, day, and distance. When I run the model: modelA<-glmmPQL(Leaving~Trial*Day*Dist,random=~1|Indiv,family="binomial") I get the error: iteration 1 Error in MEEM(object, conLin, control$niterEM) : Singularity in backsolve at level 0, block 1 >From looking at previous help

Hello, I am running polychoric correlations on a dataset composed of 12 ordinal and binary variables (N =384), using the polycor package. One of the association (between 2 dichotomous variables) is very high using the 2-step estimate (0.933 when polychoric run only between the two variables; but 0.801 when polychoric run on the 12 variables). The same correlation run with ML estimate returns a

Dear R-sians.. I am trying to plot boxplots with side-by-side option.. I tried some of the posted suggestions and could not make it work due to unequal sizes of categories... e.g. weekly measured water depth values are categorized into 5 levels based on their values such measurement is again categorized into dichotomous levels - based on the result of a test I would like generate boxplot of

Hi, I have been using some code for multivariate random forests. The output from this code is a list object with all the same values as from randomForest, but the model object is, of course, not of the class randomForest. So, I was hoping to modify the code for predict.randomForest to work for predicting the multivariate model to new data. This is my first attempt at modifying code from a

Hi everyone, Have sample of items for each one, a set of 20 dichotomous (absent-present) variables are expressed. I'm trying to understand how to explore the co-occurence of each variable. Read some papers concerning smallest space analysis, but it does not seems implemented in any R package (and my protamming skills are =0). Non metric MDS gives error messages, probably because of the

Dear R users: I have a continuous outcome variable and four predictors, two continuous and two dichotomous. i would like to use the lattice plot to create scatter plots for the continuous predictors and boxplots for the dichotomous predictors. with 4 continuous variables, this is what i have been doing: trial = rbind ( cbind ( cimt$ant.mean, cimt$age, 1 ), cbind ( cimt$ant.mean, cimt$sbp, 2 ),

I'd like to estimate an ordinal logistic regression with a random effect for a grouping variable. I do not find a pre-packaged algorithm for this. I've found methods glmmML (package: glmmML) and lmer (package: lme4) both work fine with dichotomous dependent variables. I'd like a model similar to polr (package: MASS) or lrm (package: Design) that allows random effects. I was

Hello R people, I am looking to pay someone to help write some R code. Inputs: Study identifier: ID Number for the study, each ID number is for one study only each block set should only be used for that study. This will require that you store the results from the blocks someplace on the file system. Trait #1: dichotomous rural / urban Trait #2: dichotomous sick / healthy Assignment Ratio:

Is it possible to fit a structural equation model with link functions in R? I am trying to build a logistic-regression-like model in sem, because incorporating the dichotomous variables linearly seems inappropriate. Mplus can do something similar by specifying a 'link' parameter, but I would like to be able to do it in R, ofcourse. I have explored the 'sem' package from John Fox,

Hi This is a question regarding classification performance using different methods. So far I've tried NaiveBayes (klaR package), svm (e1071) package and randomForest (randomForest). What has puzzled me is that randomForest seems to perform far better (32% classification error) than svm and NaiveBayes, which have similar classification errors (45%, 48% respectively). A similar difference in

The documentation for caret::train shows a list of parameters that one can tune for each method classification/regression method. For example, for the method randomForest one can tune mtry in the call to train. But the function call to train random forests in the original package has many other parameters, e.g. sampsize, maxnodes, etc. Is there **any** way to access these parameters using train

A few days ago I asked whether there were options other than na.action=na.fail for the R port of Breiman?s randomForest; the function?s help page did not say anything about other options. I have since discovered that a pdf document called ?The randomForest Package? and made available by Andy Liaw (who made the tool available in R---thank you) does discuss an option. It is an implementation of