04-02-2018 06:22 PM
I am using NLMIXED for Negative Binomial modeling with user defined functions. One of the models worked, but when I replaced a function from :
mu= 5*365*((MINAADT)**beta_1)* ((MAXAADT)**beta_2)*(EXP(beta_0))
5*365*(beta_0 + beta_1/((1+beta_2 *(MINAADT)**beta_3)*(1+beta_4*(MAXAADT)**beta_5)));
Following error is shown in the Log.
ERROR: QUANEW Optimization cannot be completed.
NOTE: QUANEW needs more than 200 iterations or 2000 function calls.
NOTE: PROCEDURE NLMIXED used (Total process time):
real time 0.19 seconds
cpu time 0.12 seconds
04-03-2018 04:23 PM
You didn't ask a question, but
1. If you are asking why the model didn't converge, the most likely reason is that the model does not fit the data and/or you did not provide a good initial guess for the parameters.
2. If you are asking what to do about it, then
2a. Make a better initial guess for parameters
2b. Change to a different optimize, such as TECH=NMSIMP or TECH=NRRIDG
2c.If you believe the model is almost converging, you can increase the number of iterations and the number of function evaluations that are permitted: MAXITER=500 MAXFUNC3000.