Hi
I am doing a simulation exercise using GARCH models. I have simulated data and now I want to estimate the same model with the simulated data. I have noticed that for some models the estimation results do not make much sense. I have two issues:
1. "Failure to improve likelihood after 1 iteration" => standard errors are not calculated
2. Convergence is not achieved after 500 iteration (when I set the maximum number of iteration higher it still not converge)
Are there a set of commands that could be used to delete these models?
I have some ideas:
Regards to issue number 1, I could check if the standard errors are "NA" for a specific model and then delete the model. Is this a good way to do this?
For the second issue, I could check if the number of iteration is equal to the maximum I have set.. But what is the command to get the number of iteration? Or do you know of any better solution?
Cheers
Steven
Command for number of iteration
Moderators: EViews Gareth, EViews Moderator, EViews Jason, EViews Matt
-
EViews Gareth
- Fe ddaethom, fe welon, fe amcangyfrifon
- Posts: 13604
- Joined: Tue Sep 16, 2008 5:38 pm
Re: Command for number of iteration
Unfortunately there is really no way to get the iteration number, other than freezing the output of the equation estimation and searching for that line of text. Not pretty.
Who is online
Users browsing this forum: No registered users and 2 guests
