Revisiting the CAPM model with quantile regression: creating investment strategies on the Zagreb Stock Exchange

摘要

This research explores whether conditional CAPM holds at different points of the return distribution by focusing on data from the Zagreb Stock Exchange and quantile regression methodology. Weekly data on 5 sector indices, market return on CROBEX and return on Treasury bills (91 days) for the period January 2012 to April 2018 was collected in order to empirically evaluate the CAPM model via quantile regression. The contribution of this research is given in the simulation part, where several specifications of investment strategies based on estimation results are discussed. Previous literature does not focus on utilising estimation results as guidance for dynamic investment strategies. Based upon simulations of several strategies, it was shown that quantile regression strategies could be beneficial for more conservative investors. Since this study is one of the few which try to link statistical aspects of estimating finance models with investment strategies, this research contributes to the existing literature.