Replication Data for: The Necessity of Moving Averages in Dynamic Linear Regression Models

The cumulative advice coming from the debate over lagged dependent variables in dynamic linear regression models is that including enough lags of the dependent and independent variables will fully model autocorrelation in the error term. This approach fails to account for a long-neglected source of autocorrelation in the error term: moving averages. Moving averages cannot be represented with a finite number of lags, and approximation of these terms results in either inconsistent or inefficient estimates of relevant quantities of interest, a claim we demonstrate with Monte Carlo simulations and three empirical demonstrations. We ultimately argue that moving averages should be a standard part of time series analysis and offer guidance on how to incorporate them into various modelling strategies.