Open svknair opened 3 years ago
Great suggestion. We'll think about how to incorporate them in the next version.
Best, Yiqing
On Sun, Aug 29, 2021 at 12:20 PM svknair @.***> wrote:
Hi Yiqing,
I would like to check if there is a predict option in the gsynth package. I am thinking of conducting some what-if simulations by varying the treatment time and then predicting the corresponding outcomes. For instance, what if the treatment had occurred sooner or later than the actual time. I could think of using the model output to calculate predicted values as follows:
predOutcome_t = ATT_t D_t + factor 1_t loading1 + factor2_t * loading2
- time_FE + unit_FE
I am able to obtain the loadings and unit fixed effects for each treated unit. But I am not sure how I can get average of loadings and fixed effects for all treated units. Do you have any suggestions for how the desired predicted outcomes can be calculated from the model output?
Thanks, SVK
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/xuyiqing/gsynth/issues/66, or unsubscribe https://github.com/notifications/unsubscribe-auth/AB2PKGHGHSG3FBMHVVADOGTT7KCAVANCNFSM5DARFZHQ . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
-- Yiqing Xu
Assistant Professor Department of Political Science Stanford University Web: https://yiqingxu.org/ Zoom: https://stanford.zoom.us/my/yiqingxu/
Hi Yiqing,
I would like to check if there is a predict option in the gsynth package. I am thinking of conducting some what-if simulations by varying the treatment time and then predicting the corresponding outcomes. For instance, what if the treatment had occurred sooner or later than the actual time. I could think of using the model output to calculate predicted values as follows:
predOutcome_t = ATT_t * D_t + factor 1_t * loading1 + factor2_t * loading2 + time_FE + unit_FE
I am able to obtain the loadings and unit fixed effects for each treated unit. But I am not sure how I can get average of loadings and fixed effects for all treated units. Do you have any suggestions for how the desired predicted outcomes can be calculated from the model output?
Thanks, SVK