CMStatistics 2021: Start Registration
View Submission - CFE
Title: Denoising the equity premium: A wavelet quantile approach Authors:  Ekaterini Panopoulou - University of Essex (United Kingdom) [presenting]
Antonis Alexandridis - University of Macedonia (Greece)
Abstract: The aim is to test whether it is possible to improve point, quantile and density forecasts of equity premium returns. Previous studies have shown that a variety of economic variables fail to deliver consistently accurate out-of-sample forecasts for the equity premium. We propose a novel wavelet denoising framework in the context of equity premium forecasting. First, we decompose the time-series using wavelet analysis and then, we remove the noise in each frequency using different wavelet denoising techniques. The results show that the proposed method improves the forecasting ability of linear models indicating that wavelet denoising can successfully identify the underlying persistent signal in the equity premium. Extending our framework to wavelet quantile regression, we show our approach achieves superior point, quantile and density forecasts relative to a plethora of benchmarks. Finally, our forecasting framework survives multiple-testing control. The extensive analysis and the various robustness tests indicate that the overall out-performance of our model is not an artefact of data mining.