You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi,
thank you all your efforts in developing this library
Previously, we used Google's initial R library, CausalImpact, for causal inference. Now, we're in search of a good substitute for it in Python, and this library seems like an excellent option. In the initial R library, there is functionality to extract posterior inclusion probabilities of predictors from the trained model. We use these probabilities as additional descriptive statistics to fine-tune the model and exclude certain predictors from the control group.
I'm curious if there's an approach to extract these posterior inclusion probabilities of predictors in tfcausalimpact. Your guidance on this matter would be immensely helpful.
Hi,
thank you all your efforts in developing this library
Previously, we used Google's initial R library,
CausalImpact
, for causal inference. Now, we're in search of a good substitute for it in Python, and this library seems like an excellent option. In the initial R library, there is functionality to extract posterior inclusion probabilities of predictors from the trained model. We use these probabilities as additional descriptive statistics to fine-tune the model and exclude certain predictors from the control group.I'm curious if there's an approach to extract these posterior inclusion probabilities of predictors in
tfcausalimpact
. Your guidance on this matter would be immensely helpful.Just R code to demonstrate what exactly I mean:
The text was updated successfully, but these errors were encountered: