xuyiqing / fect

fixed effect counterfactual estimators
Other
53 stars 20 forks source link

Method cfe and sfe, cfe parameters #27

Open allenzhao opened 1 year ago

allenzhao commented 1 year ago

I have a dataset that has two levels of id variables, say group_id andinvdividual_id. The Cartesian product of the two id variables can serve as the unit id as the requirement for index option.

I was wondering how can I use the method='cfe' to account for the fixed effects on group_id, 'individual_idand time? How should I specify thesfeandcfe` options?

xuyiqing commented 1 year ago

Ziyi, Do you know the answer to Allen's question?

On Tue, Mar 14, 2023 at 9:48 AM Allen Zehan Zhao @.***> wrote:

I have a dataset that has two levels of id variables, say group_id and invdividual_id. The Cartesian product of the two id variables can serve as the unit id as the requirement for index option.

I was wondering how can I use the method='cfe' to account for the fixed effects on group_id, 'individual_idand time? How should I specify thesfe andcfe` options?

— Reply to this email directly, view it on GitHub https://github.com/xuyiqing/fect/issues/27, or unsubscribe https://github.com/notifications/unsubscribe-auth/AB2PKGB7IVHUL7YZMC25WDDW4CONPANCNFSM6AAAAAAV2WM6VE . You are receiving this because you are subscribed to this thread.Message ID: @.***>

-- Yiqing Xu

Assistant Professor Department of Political Science Stanford University https://yiqingxu.org/

lzy318 commented 1 year ago

Hello Allen,

If you want to control for the group_id, invdividual_id and time but not unit id, you can use this specification:

fect(...,method = 'cfe', index = c("unit","time"), force = "time", sfe = c("group_id","individual_id") )

Best, Ziyi