[R] Several questions about MCMClogit

From: j t <jack.tmcc_at_gmail.com>
Date: Mon, 12 May 2008 15:31:50 -0500


Hello everybody,

I'm new to MCMClogit. I'm trying to use MCMClogit to fit a logistic regression model but I got some warnings I can't understand.

My input data X is 32(tissue sample)*20(genes) matrix, each element in this matrix corresponds to the expression value of one particular gene in one of 32 samples. And the Y presents the corresponding classes (0-non cancer, 1-cancer) for those 32 samples. The formula is Y~. All other parameters are default.

This is the output message:
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
The Metropolis acceptance rate for beta was 0.00000
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
Warning messages:
1: In glm.fit(x = X, y = Y, weights = weights, start = start, etastart = etastart, :
  fitted probabilities numerically 0 or 1 occurred 2: In glm.fit(x = X, y = Y, weights = weights, start = start, etastart = etastart, :
  fitted probabilities numerically 0 or 1 occurred

Question 1:
What are the meanings of two warning messages?

Question 2:
How could I use the regression coefficients to predict for other data, in other words, how could I extract those regression coefficients from the result of MCMClogit.

I know maybe my questions are some basic but it already bothered me for several days. I hope somebody can give me some hint about them. BTW, is there any specific manual or examples for using MCMClogit, the manual in package is not enough to figure out my problem.

Thanks and regards,

Chao

        [[alternative HTML version deleted]]



R-help_at_r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Received on Mon 12 May 2008 - 20:36:00 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Mon 12 May 2008 - 21:31:11 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.

list of date sections of archive