search

jmaih / RISE_toolbox

Solution and estimation of Markov Switching Rational Expectations / DSGE Models
BSD 3-Clause "New" or "Revised" License
105 stars 77 forks source link

Problem with Interfacing Dynare's csminwel1 in RISE？（2） #159

Closed wangshaobobetter closed 2 years ago

wangshaobobetter commented 3 years ago

Hi Junior,

I simplified the estimate options in “howto_rs.m" and ran. Both results are separately saved in "result_mod" and "result_rs" , but they are still quite different, such as iteration times、posterior mode、posterior stderr and log-MDD.

Before interfacing other optimizers to RISE, first I need to eliminate the confusion caused by model matching problem. Thus I hope to find the explicit issues causing the discrepancy between Dynare and RISE when using the same optimizer of fmincon.

I checked these two files for multiple times, but I still haven't find the reasons, and can't ensure the consistency. Could you please have a look at my code when you are free, it is just a small scale model. My code file is posted in the attachment.

Thank you very much!

Best,

Wang @jmaih Hansen_model.zip

jmaih commented 2 years ago

Hi Wang,

If this is still an issue, please reopen it.

Thanks