On the Existence of an Optimal Estimation Window for Risk Measures

Marcelo Brutti Righi, Paulo Sergio Ceretta
2014 Social Science Research Network  
We investigate whether there can exist an optimal estimation window for financial risk measures. Accordingly, we propose a procedure that achieves optimal estimation window by minimizing estimation bias. Using results from a Monte Carlo simulation for Value at Risk and Expected Shortfall in distinct scenarios, we conclude that the optimal length for the estimation window is not random but has very clear patterns. Our findings can contribute to the literature, as studies have typically neglected
more » ... the estimation window choice or relied on arbitrary choices.
doi:10.2139/ssrn.2487350 fatcat:k4uadp7gfrczdjcc4b4c2zalyi