Darren Callaghan wrote: > - How do you calculate initial estimates for the Beta (coefficient) > values? I've hard coded initial estimates and found values which > either always cause divergence or always converge to the same answer.
I run a discriminant analysis (a multiple linear regression with the 0/1 variable as the response). > - What's the best measure to use as a test for stopping the > iterations? I've been looking at the change in R-squared from the > linear regression result. I look at largest percent change in any of the regression coefficients. An ill-defined model might have its Rsq be stable while the coefficients were still jumping around. . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
