Affiliations: [a] Department of Mathematics and Statistics, San Diego State University, 5500 Campanile Drive, San Diego, CA 92182, USA | [b] Department of Economics, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong, China
Abstract: Hedging with financial futures is a very important tool in managing the market risk of investor portfolios. In this paper, we propose a local quadratic hedging model to estimate the risk-minimizing hedge ratios in stock futures. The hedging effectiveness variances for the local quadratic model and the conventional simple parametric model are developed, respectively. Comparisons of the within-sample and out-of-sample variances for the stock returns obtained from the Hong Kong Stock Exchange show that the local quadratic hedging model is potentially superior to the parametric method. We find that the estimated local hedge ratios are time varying which can better explain the futures market reality.
Keywords: Financial data analysis, hedging effectiveness, local quadratic estimation, within-sample and out-of-sample forecast