I've been trying a Garch(1,1) in Python using the arch package and compare the results obtained with fmincon in Matlab, so I found this article:


telling there is a difference between both estimations:

$$ \left(\omega ,\alpha,\beta \right)_{MATLAB}=\left(0.278061,0.457286,0.0328433\right) $$

$$ \left(\omega ,\alpha,\beta \right)_{PY}=\left(0.0485,5.0529e−16,0.1733\right) $$

When I try to code in my own data, I also get different results, I've also tried with R using rmgarch and rpy2, but results still differ.

  1. Is there any way to reduce this differences and get the same result from Matlab? (I can't pay the license, but need the same results)

  2. Is there any criteria to choose which language a should stick with?

Thank you in advance for your help!


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.