I need to build an S-function that lets the user pass a character vector as an argument. This function is to create data frames whose names are drawn from the character vector. I envision something l
I received helpfult replies from David L Lorenz, Patrick Burns, Julian Taylor, Barry Zajdlik, and Nick Locantore. All but one recommended the use of the assign( ) function to create dataframes within
A colleague has been trying to get SAS to fit a mixture of Weibull distributions with the numerical routines in SAS/OR. Are there any pre-existing routines available for S that fit finite mixtures of
On page 251 of "Mixed Effects Models in S and Splus," Drs. Pinheiro and Bates use the "Orthodont" dataset to illustrate the gls( ) function. I have been able to duplicate their results using Splus 6.
Last week I contacted technical support at Insightful and passed along some of my code. Splus 6.1 kept generating strange errors (among other things) when I tried to perform a stratified linear regre
Let me hasten to add that the models illustrated below were created by Insightful's tech. supt. people. They do not represent the models I was actually fitting. -david paul Last week I contacted tech
Does anyone know of a good book that demonstrates the time series methods built into Splus? I am thinking of something analogous to "Mixed-Effects Models in S and S-PLUS", by Drs. Pinheiro and Bates.
I have received several helpful replies in regards to my Time Series Book question. The replies are given below -- much thanks to Dr. Laurent Ferrara and Dr. Eric Zivot. David Paul, Ph.D. Battelle Me
My data have three columns: data$relative.humidity, data$precipitation, and data$temperature. I have been using "boxplot(split(data$relative.humidity, group), split(data$precipitation, group), split(
I have received two very helpful replies, copied below. Thanks much to Andy Liaw and Sundar Dorai-Raj. -david paul --Original Post -- My data have three columns: data$relative.humidity, data$precipit
I am aware that least squares estimates of regression coefficients in autoregressive models tend to be downward biased, particularly for small samples [Hamilton, "Time Series Analysis", 1994]. I woul