search for: gyer2

Dear R experts, I am in transition to the R system from other statistical systems I was using (SPSS, Statistica, etc.). I have used R for a few months on my Windows PC, but now I would like to install R on a Linux box with SuSE 8.1. I downloaded the RPM for SuSE 8.1 from the CRAN mirror. But when I run the INSTALL script using the midnight commander it gives the following error message: #

Dear All, I couldn't figure and couldn't google out how to make construct a pair of \Sexpr s or a LaTeX macro that would include 5\cdot 10^{-12} into the LaTeX output istead of 5e-12 . Any ideas? Thank you G?bor

Dear All! Tomorrow morning I will have to demonstrate R to a professor here at the Semmelweis University. He uses Windows and Statistica. I will have 20 minutes and I would like to convince him that R is powerful, and it even could be used in teaching the students basic medical statistics. Could you give me some ideas what to show him in this 20 minutes? This night I will have about 1.5 hours

Dear R-users, Some hours ago I successfully installed R 1.7.1 on my Linux computer (SuSE 8.1). Now I am trying to get my scripts work which I have written for R/Windows. * While with Windows I was able to use the vowels with the accents present in the Hungarian language: > d$crude.inc<-d$esetsz?m/d$?vk?zepi.l?leksz?m*1000000 > the same line leads to an error on Linux: >

Dear List, Could you tell why I get the error message? > library(multcomp) > data("cholesterol") > m = aov(response ~ trt, data = cholesterol) > cht <- glht(m, linfct = mcp(trt = "Tukey")) Error: could not find function "glht" Thank you G?bor

Dear R-help list, I have a problem with the tick marks of a Kaplan-Meier survival plot. Here is a sample: follow.up<-c(10,20,30,40,50,60,70,80,90,100) #months dead<-c(1,1,1,0,1,1,0,0,0,0) KM <-survfit(Surv(follow.up, dead)) plot(KM) The result is a nice plot. However, our research group thinks it may be a better idea to place the ticks to the years on the time scale, i.e. 0, 12, 24, 36

Dear R users, I am running a maximum likelihood model with optim. I chose the simulated annealing method (method="SANN"). SANN is not performing bad, but I guess it would be much more effecive if I could set the `parscale' parameter. The help sais: `parscale' A vector of scaling values for the parameters. Optimization is performed on `par/parscale' and these