search

ssdavenport / microsynth

Synthetic controls for micro-level data
16 stars 9 forks source link

Variance-covariance matrix from `svyglm` as model output? #25

Closed ckonichipaulo closed 2 years ago

ckonichipaulo commented 2 years ago

Hi,

First off, thanks so much for the package! I'm reaching out to see if it would be possible to add the variance-covariance matrix from the svyglm regression as part of the model output (maybe together with the other outputs in svyglm.stats?). The reason for that is that I would like to report more than one confidence interval (from permutations) without having to re-run the model again. From my understanding, the variance-covariance is used to compute delta.out and that's the only missing piece that I can't recover to build a confidence interval with a different alpha on top of the one used in the model.

Again, many thanks!

ssdavenport commented 2 years ago

Cinthia,

Thank you!

Unfortunately I don't have a lot of bandwidth to maintain the package right now, but if you wanted to push a request to the github page, I would be happy to review it and consider it for acceptance!

Steve

Sent via Superhuman ( @.*** )

On Thu, Apr 14, 2022 at 5:22 PM, Cinthia Konichi Paulo < @.*** > wrote:

Hi,

First off, thanks so much for the package! I'm reaching out to see if it would be possible to add the variance-covariance matrix from the svyglm regression as part of the model output (maybe together with the other outputs in svyglm.stats ?). The reason for that is that I would like to report more than one confidence interval (from permutations) without having to re-run the model again. From my understanding, the variance-covariance is used to compute delta.out and that's the only missing piece that I can't recover to build a confidence interval with a different alpha on top of the one used in the model.

Again, many thanks!

— Reply to this email directly, view it on GitHub ( https://github.com/ssdavenport/microsynth/issues/25 ) , or unsubscribe ( https://github.com/notifications/unsubscribe-auth/AAGG6RVQDQEO3EWMB564773VFCEC5ANCNFSM5TO5PNBA ). You are receiving this because you are subscribed to this thread. Message ID: <ssdavenport/microsynth/issues/25 @ github. com>

ckonichipaulo commented 2 years ago

Hi Steven!

Thank you for your reply! I created a PR but this is my first time ever contributing to a package, so please let me know if there is anything else that I should add to it.

Thanks, Cinthia

ssdavenport commented 2 years ago

Awesome! thank you for your contribution Cinthia

Sent via Superhuman ( @.*** )

On Mon, Apr 18, 2022 at 7:25 PM, Cinthia Konichi Paulo < @.*** > wrote:

Hi Steven!

Thank you for your reply! I created a PR ( https://github.com/ssdavenport/microsynth/pull/26 ) but this is my first time ever contributing to a package, so please let me know if there is anything else that I should add to it.

Thanks, Cinthia

— Reply to this email directly, view it on GitHub ( https://github.com/ssdavenport/microsynth/issues/25#issuecomment-1101850510 ) , or unsubscribe ( https://github.com/notifications/unsubscribe-auth/AAGG6RUFRQ2VUAHJZBMMG53VFXVOFANCNFSM5TO5PNBA ). You are receiving this because you commented. Message ID: <ssdavenport/microsynth/issues/25/1101850510 @ github. com>