Dealing with log-binomial convergence problem
Hi everyone 
I am looking for a way to change the "weight function of pi" [1/(1-Pi)] in the GLM model with binomial distribution and log link.
One of the suggested approaches to avoid convergence problems is to use Maclaurine series as the weight function of Pi.
I wonder if there is a user-defined option for the weight function in the GLM command?
I appreciate your advice in advance.
I am looking for a way to change the "weight function of pi" [1/(1-Pi)] in the GLM model with binomial distribution and log link.
One of the suggested approaches to avoid convergence problems is to use Maclaurine series as the weight function of Pi.
I wonder if there is a user-defined option for the weight function in the GLM command?
I appreciate your advice in advance.
