Estimating heterogeneous costs of participation in the risky asset markets

This paper develops and estimates a dynamic structural model of participation in the risky financial asset markets using household level panel data. We specify a simple economic model in order to capture the portfolio choice over the life cycle. We solve the model using numerical techniques. Then we embed the optimal solution into the statistical (auxiliary) model and estimate the structural parameters using Generalized Indirect Inference. This paper focuses on the estimation of the non proportional costs to participate in the risky asset markets. We consider heterogeneous costs among education groups. We find that participation costs in the risky asset markets are positive and significant. We also conclude that they vary a lot among education groups.