similar to: How to do leave-n-out cross validation in R?

Hi, I was trying to do cross-validation using the crossval function (bootstrap package), with the following code: --------------------------------------------------------------------------------------------------------- theta.fit <- function(x,y){ model <- svm(x,y,kernel = "linear") } theta.predict <- function(fit,x){ prediction <- predict(fit,x)

The function crossval (in library bootstrap) works well for the first degree polynomial model, but in the case of the second degree model I got an error message (see below). I would be very greatfull if somebody could give some advices for the following: > library(bootstrap) > > x<-c(22,23.4,24.9,28.5,29.8,31.6,34.2,36.4,37.7,39) >

I'm seeing some strange behavior while using the snow package for networked computers. I believe it's caused by name space resolution issues, and would appreciate any suggestions tracking it down. First, is there a way to find out what frame (as in frames in environments, not data frames) a name is being obtained from or put into? Second, how closely does the evaluation environment in

Hi, Newbie question about the pls package. Setup: Mac OS 10.3.9 R: Aqua GUI 1.01, v 2.0.1 I want to get R^2 and Q^2 (LOO and Leave-10-Out) values for each component for my model. I was running into a few problems so I played with the example a little and the results do not match up with the comments in the help pages. $ library(pls) $ data(NIR) $ testing.plsNOCV <- plsr(y ~ X, 6, data =

Using R 1.7.0 I get * checking parcv-manual.tex ... ERROR Could not create DVI version. Although there is no apparent error. The dvi file exists. Possibly there is some problem with my TeX setup, but the following messages don't suggest that either. Here's the full log, which does show some documentation issues: sheep:~$R CMD check --library=.R/library/ src/parcv * checking for working

Hello r-help, I am trying to collapse or aggregate 'some' of a data frame. A very simplified version of my data frame looks like: > tester trip set num sex lfs1 lfs2 1 313 15 5 M 2 3 2 313 15 3 F 1 2 3 313 17 1 M 0 1 4 313 17 2 F 1 1 5 313 17 1 U 1 0 And I want to omit sex from the picture and just get an addition of num,

summary: The function 'by' inconsistently strips class from the data to which it is applied. quick reason: tapply strips class when simplify is set to TRUE (the default) due to the class stripping behaviour of unlist. quick answer: This can be fixed by invoking tapply with simplify=FALSE, or changing tapply to use do.call(c instead of unlist executable example:

Dear List, I have a logical data frame with NA's and a grouping factor, and I want to calculate the % TRUE per column and group. With an indexed database, result are mainly limited by printout time, but my R-solution below let's me wait (there are about 10* cases in the real data set). Any suggestions to speed this up? Yes, I could wait for the result in real life, but just curious if I

Hi   I want to do leave-one-out cross-validation for multinomial logistic regression in R. I did multinomial logistic reg. by package nnet in R. How I do validation? by which function? response variable has 7 levels   please help me   Thanks alot Azam [[alternative HTML version deleted]]

Hi all, Has anyone tried to do k-fold cross validation for flexible discriminant analysis ( mda library), for example, using crossval() in bootstrap? The problem is that the function crossval() requires a separate matrix for predictors and another for responses, whereas the function fda(), using the formula argument only. Is there another way of doing k-fold cross validation for functions which

Greetings, I have been exploring the use of the caret package to conduct some plsda modeling. Previously, I have come across methods that result in a R2 and Q2 for the model. Using the 'iris' data set, I wanted to see if I could accomplish this with the caret package. I use the following code: library(caret) data(iris) #needed to convert to numeric in order to do regression #I

I would like to put confidence intervals on a mantel corellogram they are already calculated in the pmgram object but I am unsure how I get the x value in order to plot them? package(ecodist) X<-1:100 Y<-rnorm(1:100) Z<-rnorm(1:100) XY<-dist(data.frame(X,Y)) YX<-dist(data.frame(Y,X)) my.mgram<-mgram(XY,XZ) plot(my.mgram) print(my.mgram) > print(my.mgram) $mgram

Howdy R gurus ! I am newbie to R I use R 2.0.1 in Windows XP. When I run R I got the follwoing memory error. My physical memory size is 3 Gb. My R got the memory problem when it reached to about 2 Gb. Thanks in advance, > library(spdep) > sfr.lagsarlm <- lagsarlm(sfr.data$Bldgsqft ~ sfr.data$Ncounty + sfr.data$Nugb + sfr.data$Ngroup, data=sfr.data, listw=sfr.listw,

Hi, I'm running Samba 3.0.11 on RedHat ES 3 kernel version 2.4.21-15.0.4.ELsmp and have a quick question about AD group membership limits Am I right in assuming that Samba is limited by the group membership parameters (ie NGROUP = 32) imposed by the Linux kernel? Is there any workaround in Samba for this? At the moment if a user is a member of more then 32 domain groups they cannot access

Hi, I am trying to make a palaeoenvironmental transfer function using the R package rioja that predicts the water-table (measured as depth to the water table) of an area given the testate amoebae that are found there. I've carried out weighted averaging of the data and am trying to produce a graph that shows the observed water-table versues the model's predicted values. Following the

Dear R helpers I'm attempting to create a matrix of scatterplots with X-Y mean lines and regression lines added to each plot in the matrix. I have managed to create the first plot I would like using scatterplot but have run into an odd result when I use par() to set up the page to take multiple plot. Specifically, the mean and regression lines appear to plot in the second plot window, not

Dear R users, I am sorry to disturb you! But I really need your help for the usage of sigPathwy. Actually, I want a sliding window analysis for possible chromosome expression pattern mining. My research microorganism is a plant pathogen, Gibberella zeae, and I first used SAS to divide locus number with 10, 20, 30, or 40 on the fungal chromosome according to their location. I really

A few problems in R-0.50-a3 (which were also in R-0.49): 1) y <- c(10,11,12,13,14,NA,NA) n <- length(y) missed <- (1:n)[is.na(y)] notmissed <- (1:n)[!is.na(y)] blocks <- cut(missed,breaks=c(0,notmissed,n+1)) a <- function(v) { q <- range(v) c(q[1]-1,q[2]+1) } brackets <- tapply(missed,blocks,a) This codes gives the following in S: > brackets $"0+ thru 1":

Hi folks, I'm trying to estimate bias-corrected percentile (BCP) confidence intervals on a vector from a simple for loop used for resampling. I am attempting to follow steps in Manly, B. 1998. Randomization, bootstrap and monte carlo methods in biology. 2nd edition., p. 48. PDF of the approach/steps should be available here: https://wyocoopunit.box.com/s/9vm4vgmbx5h7um809bvg6u7wr392v6i9 If

Hi- I want to fit a model with crossed random effects and heteroskedastic level-1 errors where inferences about fixed effects are of primary interest. The dimension of the random effects is making the model computationally prohibitive using lme() where I could model the heteroskedasticity with the "weights" argument. I am aware that the weights argument to lmer() cannot be used to