On Sun, 18 Jan 2004 20:38:50 -0800, Koh Puay Ping wrote: > HI, > > Can you recommend any good method to choose the best model for a 12 > variable logistic model? Should i always try to get a model with very > good fit (H & L goodness of test using SAS)?
See if you can get a hold of a copy of Frank Harrell's book "Regression Modeling Strategies", which covers this subject in great detail. It would hopefully be available from a university main or departmental library in your area. An Amazon.com link for the book is: http://www.amazon.com/exec/obidos/tg/detail/-/0387952322 HTH, Marc Schwartz . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
