Flexible conditional density estimation for time series
This provides a more accurate tool for time series analysis, but it is incremental as it builds on existing regression-based density estimation methods.
The paper tackles the problem of conditional density estimation for time series by introducing FlexCodeTS, a flexible nonparametric estimator that adapts to data structure using arbitrary regression methods, and it generally outperforms NNKCDE and GARCH in simulations and real data based on CDE and pinball loss metrics.
This paper introduces FlexCodeTS, a new conditional density estimator for time series. FlexCodeTS is a flexible nonparametric conditional density estimator, which can be based on an arbitrary regression method. It is shown that FlexCodeTS inherits the rate of convergence of the chosen regression method. Hence, FlexCodeTS can adapt its convergence by employing the regression method that best fits the structure of data. From an empirical perspective, FlexCodeTS is compared to NNKCDE and GARCH in both simulated and real data. FlexCodeTS is shown to generally obtain the best performance among the selected methods according to either the CDE loss or the pinball loss.