Locally Stationary Quantile Regression for Inflation and Interest Rates

• Activity Number: 27
• Type: Contributed
• Date/Time: Sunday, August 3, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Section on Nonparametric Statistics
• Author(s): Seonjin Kim*+ and Zhibiao Zhao
• Companies: Miami University and Penn State
• Keywords: Exogenous autoregressive model ; Locally stationary models ; Quantile regression ; Stock return-inflation puzzle ; Nonparametric inferences ; Time-varying models

Abstract

We derive a time-varying locally stationary quantile regression model for conditional quantile regression of inflation on the past inflation and interest rates, with the coefficient varying with time nonparametrically. Also, we study nonparametric estimations for these time-varying and quantile specific coefficients and establish their asymptotic theory. To improve finite sample performance, a wild bootstrap assisted inference is introduced. In empirical application, our model reveals interesting features which cannot be captured by least-squares regression.