I have also received this message: Failed: Cannot Decrease Objective Function
when running a logistic regression. I am not sure what this means or how to fix this problem; I've run many logistic regressions and never received this error message before.
Thanks for any insight you might have!
The Nonlinear platform is very powerful and versatile. It has many options. Therefore, there are many reasons why you might get such an error message. The objective function, which could be many different functions, is monitored as the parameter estimates are updated. Since the function did not change (decrease), it cannot tell if it converged. So it would help if you could describe the nature of your
Otherwise it is difficult to provide you a useful answer.
Did you try one of the suggestions? In particular, did you accept the current estimates and examine the fitted model to see if it was acceptable?