Dr Ekaterini Panopoulou from Kent Business School will present a working paper entitled “Out-of-sample equity premium prediction: a complete subset quantile regression approach” at the 7th MIFN conference to be held at the University of Namur, Belgium between the 23rd and 24th September 2013.
The conference is designed to facilitate high level, high intense discussion and fruitful exchange between researchers both in the domain of international finance and/or econometrics of finance. Plenary speakers include Prof. Charles Goodhart (London School of Economics) and Prof. Peter Schotman (University of Maastricht).
Dr Ekaterini Panopoulou from Kent Business School will present a working paper entitled “Out-of-sample equity premium prediction: a complete subset quantile regression approach”
This paper proposes a quantile regression approach to equity premium forecasting based on complete subset combinations of forecasts. The proposed approach extends the complete subset mean regression framework to a quantile regression setting. This framework enables us to construct robust and accurate equity premium predictions from a set of complete subset quantile forecasts. Finally, we put forward a recursive algorithm that selects, in real time, the best complete subset for each predictive regression quantile. Robust point forecasts of the equity premium are based on the synthesis of these ‘optimal’ complete subset quantile forecasts. The proposed approach delivers statistically and economically significant out-of-sample forecasts relative to both the historical average benchmark and the complete subset mean regression approach.