Closed flh3 closed 1 year ago
@flh3 the standard errors on the coefficients are always CR-0 in WeMix. I agree the documentation needs to be updated.
fixed. Please confirm I attributed this correctly
Looks good! Thanks.
From: Paul Bailey @.> Sent: Thursday, May 18, 2023 6:00 PM To: American-Institutes-for-Research/WeMix @.> Cc: Huang, Francis Howard L. @.>; Mention @.> Subject: Re: [American-Institutes-for-Research/WeMix] robust standard errors? (Issue #9)
WARNING: This message has originated from an External Source. This may be a phishing expedition that can result in unauthorized access to our IT System. Please use proper judgment and caution when opening attachments, clicking links, or responding to this email.
fixed. Please confirm I attributed this correctlyhttps://github.com/American-Institutes-for-Research/WeMix/blob/main/inst/NEWS.Rd#L7
- Reply to this email directly, view it on GitHubhttps://github.com/American-Institutes-for-Research/WeMix/issues/9#issuecomment-1553759650, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AB3SG54VIXU53H56EJ5HKHDXG2SXBANCNFSM6AAAAAAYFM5HYI. You are receiving this because you were mentioned.Message ID: @.***>
Hello,
Just wanted to confirm: are standard errors (SEs) used in mix robust SEs? The mix documentation indicates:
"the standard errors of the fixed effects, robust if robustSE was set to true"
However, there is no mention of where to set robustSE and what is the default. I am asking since at times, I would like to compare the model-based vs. robust standard errors.
Thank you.