A simple nonparametric conditional quantile estimator for time series with thin tails

In this study, we consider a simple conditional quantile estimator in a nonparametric framework with time series data. We prove the consistency and asymptotic normality of our simple estimator for absolutely regular processes ($\beta$-mixing). This simple estimator can get better finite sample performances at thin tails than the check-function-based estimator. Finite sample simulation results show that our simple estimators have better finite sample performances at thin tails of time series data.