Hi! I'm using the user-defined optimization feature of Eviews, and although the code run well, it sometimes ends with the warning "Optimization may be unreliable (first or second order conditions not met)". What does this really mean and any suggestion on how to fix it? I've tried different starting values and optimization method (BFGS, legacy, ...) but still got this warning most of the time. The optimization equation I'm using has many regresssors and is a bit complicated, but I wonder if that's the whole story. Thanks!
↧