[R] ordinal regression

[Frank E Harrell Jr]

Iasonas Lamprianou wrote:
> Dear colleagues,
> 
> i am carrying out an ordinal regression model. I try it on SPSS but I "flirt" with R as well. I have a few questions.
> 
> 1. What is the most reliable/tested/trusted package for ordinal regression in the R world? 
> 
> 2. Also, I have a statistical question. What is the danger of having to many 'empty cells' in ordinal regression? How many empty cells are too many? Do you have a reference for me to read?
> 
> Thank you for the support
> 
> 
> Dr. Iasonas Lamprianou

Not sure what is the most trusted but the lrm function in the rms 
package does proportional odds and continuation ratio models (the latter 
with a trick to expand the dataset).  Nearly empty cells are not a 
problem unless you allow for non-proportionality (e.g., interaction 
between X and Y).

Frank

> 
> 
> Assistant Professor (Educational Research and Evaluation)
> Department of Education Sciences
> European University-Cyprus
> P.O. Box 22006
> 1516 Nicosia
> Cyprus 
> Tel.: +357-22-713178
> Fax: +357-22-590539
> 
> 
> Honorary Research Fellow
> Department of Education
> The University of Manchester
> Oxford Road, Manchester M13 9PL, UK
> Tel. 0044  161 275 3485
> iasonas.lamprianou at manchester.ac.uk
> 
> 
> 


-- 
Frank E Harrell Jr   Professor and Chairman        School of Medicine
                      Department of Biostatistics   Vanderbilt University