search

xuyiqing / gsynth

Generalized Synthetic Control Method
Other
134 stars 41 forks source link

wgt.implied becomes "NULL" #17

Open Nakatage opened 6 years ago

Nakatage commented 6 years ago

Sometimes, I got the NULL from wgt.implied. what is the cause of this?

hshieh commented 5 years ago

I am having the same issue.

xuyiqing commented 5 years ago

We'll look into it. Thanks, guys!

On Fri, Aug 16, 2019 at 11:36 AM hshieh notifications@github.com wrote:

I am having the same issue.

— 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/17?email_source=notifications&email_token=AB2PKGGYXK4HRIHY34FNLDLQE3CPTA5CNFSM4FCLKWJKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD4O6KHY#issuecomment-522052895, or mute the thread https://github.com/notifications/unsubscribe-auth/AB2PKGHZFGDPQ5ONWSYKFJDQE3CPTANCNFSM4FCLKWJA .

-- Yiqing Xu

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

liulch commented 5 years ago

Sometimes, I got the NULL from wgt.implied. what is the cause of this?

I just checked the code and found that there are two possible reasons: the first one is that there is no factors are selected so there are no implied weights. The second one is that the generalized inverse matrix that represents the implied weights is not solvable so that the function returns a NULL.

hshieh commented 5 years ago

Thanks, Licheng. What I actually want to do is to show the (weighted) average characteristics/outcomes of the synthetic unit side-by-side with the characteristics/outcomes of the treated unit. Is there a way to do that without the implied weights? It seems my estimations on my data sample does not yield a solvable matrix. Thanks for your help and work on this package.