Quantile Regression For Panel Vector Auto Regression Model

This paper studies parameter estimate problem of the quantile regression for panel data vector auto regression model. Panel data fixed effects estimators are typically biased when N is larger. To reduce the bias, we suggest the use of the instrumental variables quantile regression method of Chernozhukov(2006). Monte Carlo simulation shows that the instrumental variables approach sharply reduces the bias when N is larger than T. Finally, we illustrate a real data example to study the relationship between the real estate prices and macroeconomic of35large and medium-sized cities.