search

google / CausalImpact

An R package for causal inference in time series
Apache License 2.0
1.71k stars 254 forks source link

handle past interventions #25

Open rquintino opened 5 years ago

rquintino commented 5 years ago

Hi! what is the best way to handle past known interventions effects in the time series? thx!

AlexSiormpas commented 5 years ago

Maybe with a dummy predictor variable that takes 1 during the dates of the older intervention and 0 for all the other dates