Estimates given by the lowest SSR found by grid search on a parameter

For questions regarding programming in the EViews programming language.

Moderators: EViews Gareth, EViews Jason, EViews Moderator, EViews Matt

remuct
Posts: 27
Joined: Fri May 12, 2017 6:47 am

Estimates given by the lowest SSR found by grid search on a parameter

Postby remuct » Tue Feb 01, 2022 8:00 pm

Hello,
I am estimating a model where the optimal value of one parameter (say beta) is determined using a grid search over a given interval and for which the SSR is minimized. What I am doing is the following: a) I run a regression for each value of beta in the interval using a for next loop and calculate the corresponding SSR; b) I store all the obtained SSRs in a vector; c) I find the lowest element of this vector, that is, the lowest SSR; d) then I rerun my regressions by grid searching over beta (using a new for next loop) and comparing (using an if statement) for each value of beta the SSR with the minimum SSR I found previously; e) when the current SSR matches with the minimum SSR then I exit from the for next loop and I have my optimal estimates and the regression output. .
This approach works well but is a rather heavy to implement. Is there a more easy way to find the vector of estimates corresponding to the lowest SSR found by a grid search over a parameter, a built-in function which does directly the job?
Thanks

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13294
Joined: Tue Sep 16, 2008 5:38 pm

Re: Estimates given by the lowest SSR found by grid search on a parameter

Postby EViews Gareth » Tue Feb 01, 2022 8:48 pm

Without knowing the functional form of beta, no, not with a grid search.
Follow us on Twitter @IHSEViews


Return to “Programming”

Who is online

Users browsing this forum: No registered users and 19 guests