The literature on agent-based models has been highly successful in replicating many stylized facts of financial and macroeconomic time series. Over the past decade, however, also advances in the estimation of such models have been made. Due to the inherent heterogeneity of agents and nonlinearity of agent-based models, fundamental choices have to be made to take the models to the data. In this chapter we provide an overview of the current literature on the empirical validation of agent-based models. We discuss potential lessons from other fields of applications of agent-based models, avenues for estimation of reduced form and 'full-fledged' agent-based models, estimation methods, as well as applications and results.