search

insongkim / PanelMatch

111 stars 34 forks source link

Applying PanelMatch to zero-inflated count data as DV? #102

Closed LarsLott closed 2 years ago

LarsLott commented 2 years ago

Dear @adamrauh @kosukeimai,

thanks for this really powerful R package! Is it possible to have a (zero-inflated) count variable as an outcome variable instead of a continuous one? Does the application of zero-inflated count data as an outcome variable contradict any of PanelMatch´s assumptions? I am currently working on how autocratiaztion onset (treatment) affect domestic terrorism (count data that is zero-inflated).

Thanks for your clarification. Best, Lars

kosukeimai commented 2 years ago

@LarsPelke Yes the outcome can be anything. For your case, you can interpret the effect to be the average number of terrorism events caused by autocritization.