Displaying 1 result from an estimated 1 matches for "pneumo2".
2007 Jan 06
2
Using VGAM's vglm function for ordinal logistic regression
...concerns the format in which the response variable must be provided for
this function to work correctly. Consider the following example:
------
library(VGAM)
library(MASS)
attach(pneumo)
pneumo # Inspect the format of the original dataset
# Restructure the pneumo dataset into a different format
pneumo2 <- data.frame(matrix(ncol=3, nrow=24))
colnames(pneumo2) <- c('exposure.time', 'severity', 'freq')
pneumo2[,1] <- rep(pneumo[,1],3)
pneumo2[,2] <-
as.ordered(c(rep('normal',8),rep('mild',8),rep('severe',8)))
pneumo2[1:8,3] <- pneumo[...