Newey west how many lags

Shivam: You have posted in essence two reminders "What about my question? As you are relatively new to the list, I was giving a hint on how it does and unfortunately in this case for you so far does not work.

The absence of an answer suggests that people didn't find the question clear or that it is just too recondite to answer. We've occasionally had people confuse Statalist with the company's technical support, which is under some obligation to answer, even if the answer is some version of "Sorry, no".

You've clarified that optimal lag would mean optimising some IC. I don't use newey but I guess that what you want would require some programming from you, not only to calculate your favourite IC, but also to loop over possible lags. It doesn't seem provided for in the official command. You may get a much better answer from someone closer to your field than I am. Thanks for the hint Nick. I will keep this in mind. Regarding optimal lag, I was hoping the official command may be able to incorporate this, as varsoc is able to decide optimal lags for various criterion.

It seems this is not the case. I will wait for someone more knowledgable to be kind enough to answer. Attaullah Shah. Regards Attaullah Shah, PhD. Thanks Attaullah, I will investigate further. Edit: I've looked at the link and installed the packages and followed the commands. I think this is exactly what I'm trying to do, however I'm not confident. Are you familiar with this procedure? Can you confirm that it performs OLS regression with Newey-West robust standard errors, automatically choosing an optimal lag, for the regression of one y variable and three x variables which I put in.

Last edited by Shivam Patel ; 27 Jul , Jorge Eduardo Perez Perez. As Attaullah pointed out, the trick is to use the user written command ivreg2 which includes Newey West standard errors with automatic banwidth selection. Thank you so much Jorge! You are literally my hero, once again. I will investigate and report back my findings. It works like a charm. I've been doing some reading, and it seems P-values will be important for my analysis as they identify whether a factor is significant in explaining the returns.

How do I generate p-values? I've found an online tool that can create them using t-stats, however I'm unsure as to what degrees of freedom would be. If I can do it in Stata, that'd be perfect. Because the sample is greater than say, 30? Last edited by Shivam Patel ; 01 Aug , This example might help Code:.

Previous 1 2 Next. Yes No. My answer is going to expand on what Achim mentioned as "the growth rate of this lag length parameter". These are obviously not the only conditions for the consistency of the estimators, but the other conditions are not directly related to the lag length. Greene Econometric Analysis, 7th edition, section Disclaimer: I do not know the details about the panel version of the Newey-West standard errors and their implementation in Stata.

The references I give below are for time series data without panels but hopefully these take you a step further. Some implementations do use lag selection heuristics based on this growth rate, though. Sign up to join this community.

The best answers are voted up and rise to the top. Stack Overflow for Teams — Collaborate and share knowledge with a private group. Create a free Team What is Teams? Learn more. Asked 6 years, 1 month ago.

Active 2 years, 4 months ago. Viewed 14k times. Any help would be appreciated! Improve this question. Bass C. Bass 41 1 1 gold badge 1 1 silver badge 2 2 bronze badges. Eviews has a built-in function to do this. Add a comment. Active Oldest Votes. Improve this answer. Candamir Candamir 8 8 silver badges 27 27 bronze badges.

Achim Zeileis Achim Zeileis