0
$\begingroup$

I am wondering what type of covariance correction for standard errors is better: Hansen-Hodrick or Newey-West?

Also, does someone know if StatsModels package that uses "HAC" for robust covariance uses Hansen-Hodrick or Newey-West correction for standard errors?

statsmodels.regression.linear_model.RegressionResults.get_robustcov_results
$\endgroup$
1
$\begingroup$

statsmodels uses by default Newey-West corrected standard errors with the usual Bartlett window.

There is a 'weights_func' or 'kernel' option to choose a different window than Bartlett, eg. uniform.

However, statsmodels has no other options for HAC robust standard errors like pre-whitening or automatic lag selection, or autocorrelation robust standard errors without heteroscedasticity robustness (i.e. only 'HAC', but no AC).

If I remember correctly Hansen-Hodrick is 'AC' with uniform kernel.

$\endgroup$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.