Combining counterfactual outcomes and ARIMA models for policy evaluation

Summary The Rubin Causal Model (RCM) is a framework that allows to define the causal effect of an intervention as contrast potential outcomes. In recent years, several methods have been developed under RCM estimate effects in time series settings. None these makes use autoregressive integrated moving average (ARIMA) models, which are instead very common econometrics literature. this paper, we propose novel approach, named Causal-ARIMA (C-ARIMA), and observational settings RCM. We first formalise assumptions enabling definition, estimation attribution intervention. then check validity proposed method with simulation study. empirical application, C-ARIMA assess permanent price reduction on supermarket sales. CausalArima R package provides implementation approach.