City Research Online

Abstract

The increasing availability of intraday financial data has led to improvements in daily volatility forecasting through long-memory models of realized volatility. This paper demonstrates the merit of the non-parametric Nearest Neighbor (NN) approach for S&P 100 realized variance forecasting. A priori the NN approach is appealing because it can reproduce complex dynamic dependencies while largely avoiding misspecification and parameter estimation uncertainty, unlike model-based methods. We evaluate the forecasts through straddle trading profitability metrics and using conventional statistical accuracy criteria. The ranking of individual forecasts confirms that statistical accuracy does not have a one-to-one mapping into profitability. In turbulent markets, the NN forecasts lead to higher risk-adjusted profitability even though the model-based forecasts are statistically superior. In both calm and turbulent market conditions, the directional combination of NN and model-based forecasts is more profitable than any of the individual forecasts.

Publication Type:	Article
Additional Information:	© 2015, Elsevier. Licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International http://creativecommons.org/licenses/by-nc-nd/4.0/
Publisher Keywords:	Realized Volatility; Volatility Forecasting; Non-parametric Forecasts; Nearest Neighbor; Long-Memory Models; Forecast Combination; Straddles; Options Trading
Subjects:	H Social Sciences > HG Finance Q Science > QA Mathematics
Departments:	Bayes Business School > Finance

Export

Downloads