s-news
[Top] [All Lists]

ordered logistic regression

To: s-news@lists.biostat.wustl.edu
Subject: ordered logistic regression
From: Markus Haverkamp <markus.haverkamp@algorithmics.com>
Date: Wed, 11 Apr 2007 14:57:44 +0100

Dear Ladies and Gentlemen,

 

I am new to the Mailing list and have a question concerning the ordered logistic regression command polr in the mass library:

 

My script imports data from an Excel file and stores it in a variable called “Dataset”:

Dataset <- importData(“file.xls”, startCol = 1, pageNumber = 4, type = "EXCEL")

 

The formulas for the ordered logistic regression are stored in an array:

modelformula <- c("formula1”,”formula2”,…)

 

The script runs through a for loop and calculates one formula in the array after the other. The coefficients, errors and t-values of each calculation are supposed to be stored in a row of the two dimensional array “results”:

for (count in 1:length(modelformula)){

   mod<-polr(formula = modelformula[count], data="" na.action = "">

   results[count,5:7]<-coefficients(summary(mod))[1,1:3]

}

 

For the first iteration the array “results” contains the correct results. However, for the second and all further iterations only the coefficients stored in “results” seam correct, and the standard errors and t-values are not. In fact they are quite similar to the standard errors and t-values from the first iteration.

 

Does somebody know why I do not get the correct standard errors and t-values as from the second iteration? Is this a bug in S-Plus or do I have to resolve an issue with Pointers?

 

Best regards,

 

Markus Haverkamp

 

S-Plus version: S-PLUS® 7.0 for Windows PROFESSIONAL DEVELOPER

Windows version: Microsoft Windows XP Professional Version 2002 Service Pack 2

 


This email and any files transmitted with it are confidential and proprietary to Algorithmics Incorporated and its affiliates ("Algorithmics"). If received in error, use is prohibited. Please destroy, and notify sender. Sender does not waive confidentiality or privilege. Internet communications cannot be guaranteed to be timely, secure, error or virus-free. Algorithmics does not accept liability for any errors or omissions. Any commitment intended to bind Algorithmics must be reduced to writing and signed by an authorized signatory.

<Prev in Thread] Current Thread [Next in Thread>
  • ordered logistic regression, Markus Haverkamp <=