Methods for policy evaluation with panel data

The focus of this work is present some novel developments in public policy evaluation methods. All four works that make up this dissertation are described under the Rubin causal model (RCM), a framework to estimate causal quantities, under some realistic assumptions and using some of the recently developed estimators. I focused my attention on application to structured data, such as panel data, spatial data and time series. The dissertation is structured as follows: the first chapter introduces the use of the Synthetic Control Method for policy evaluation under a partial interference framework. The second chapter presents the evaluation of conditional cash lotteries, implemented in the US, to foster Covid-19 vaccination. The third chapter presents a novel estimator for causal quantities in presence of spatially correlated treated units. The last chapter deals with a novel application of the principal stratification method to evaluate an active labour market policy.