Unified Inference for Dynamic Quantile Predictive Regression

TL;DR

This paper introduces a unified asymptotic distribution framework for dynamic quantile predictive regressions, enabling analysis of stock return predictability even with nonstationary data.

Contribution

It provides a novel theoretical approach to handle nonstationarity in quantile predictive regressions for financial data.

Findings

Unified asymptotic theory developed

Applicable to nonstationary stock return data

Enhances understanding of quantile predictability

Abstract

This paper develops unified asymptotic distribution theory for dynamic quantile predictive regressions which is useful when examining quantile predictability in stock returns under possible presence of nonstationarity.