A0461
Title: Panel quantile regression for extreme risk
Authors: Xuan Leng - Xiamen University (China) [presenting]
Abstract: Panel quantile regression models play an essential role in finance, insurance, and risk management applications. However, a direct application of panel regression for the extreme conditional quantiles may suffer from significant estimation errors due to data sparsity on the far tail. A two-stage method is introduced to predict extreme conditional quantiles over cross-sections. First, use panel quantile regression at a selected intermediate level, then extrapolate the intermediate level to an extreme level with extreme value theory. The combination of panel quantile regression at an intermediate level and extreme value theory relies on a set of second-order conditions for heteroscedastic extremes. Also, a metric called Average Absolute Relative Error is proposed to evaluate the prediction performance of both intermediate and extreme conditional quantiles. Individual fixed effects in panel quantile regressions complicate the asymptotic analysis of the two-stage method and prediction metric. Compared to the direct use of panel quantile regression, the finite sample performance of the extreme conditional quantile prediction compared is demonstrated. Finally, the two-stage method is applied to the macroeconomic and housing price data, and strong evidence of housing bubbles and common economic factors is found.
