Enhancing hedging performance with the spanning polynomial projection |
| |
Authors: | An-Sing Chen Yan-Zhen Liu |
| |
Affiliation: | 1. Department of Finance , National Chung Cheng University , 168 University Rd., Min-Hsiung, Chia-Yi 621, Taiwan, R.O.C finasc@ccu.edu.tw;3. Department of Finance , National Chung Cheng University , 168 University Rd., Min-Hsiung, Chia-Yi 621, Taiwan, R.O.C |
| |
Abstract: | Statistical time-series approaches to hedging are difficult to beat, especially out-of-sample, and are capable of out-performing many theory-based derivative pricing model approaches to hedging commodity price risks using futures contracts. However, the vast majority of time-series approaches to hedging discussed in the literature are essentially linear statistical projections, whether univariate or multivariate. Little is known about the potential hedging capabilities of nonlinear methods. This study describes how least-squares orthogonal polynomial approximation methods based on the spanning polynomial projection (SPP) can be used to enhance standard (linear) optimal hedging methods and improve hedging performance for a hedger with a mean–variance objective. Empirical analyses show that the SPP can be used effectively for hedging and gives better out-of-sample hedging performance than the benchmark VEC-GARCH hedging model. Results are robust to the inclusion of transaction costs and risk-aversion assumptions. |
| |
Keywords: | Hedging with utility-based preferences Implementation of optimal hedging Derivatives hedging Hedging techniques Hedging errors Trading costs Forecasting ability Forecasting applications |
|
|