The Estimation And Comparison Research Of Optimal Hedge Ratio For Commodity Futures

The essential function of futures markets is hedge and the core of hedge is how to estimate optimal hedging ratio. Objective function, linkage mode of futures price and spot price, and estimate method constitutes three-dimensional space of hedge. This study focus on the estimation and comparison of optimal hedging ratio for commodity futures from those three angle.The first chapter is exordium and is about significance of research, present research review, research trend, technology courses and innovation of this article.The second chapter propose random coefficient Markov regime switching model to estimate optimal hedge ratio of china's copper futures market. RCMRS model jumps out the research framework of the GARCH model which bases on the conditional second moments, treats the optimal hedge ratio as random coefficient and estimate time-varying hedge ratio that are regime dependent directly. Regime states are treated as latent variables and estimated along with the other parameters of RCMRS model using maximum likelihood techniques. By allowing the volatility to switch stochastically between different processes under different market conditions, the RCMRS hedge ratios exhibit lower variability compared to the GARCH hedge ratio. The hedging performance of RCMRS model is compared against GARCH, VECM, VAR and OLS model using the minimum variance and maximum utility approaches over both an ex post and ex ante hedge period. RCMRS model outperforms the other hedging strategies within sample. In the out-of-sample analysis however, the RCMRS model is outperformed by the other models.Based on kalman filter, the third chapter use state space model to estimate time varying optimal hedge ratio of china's copper futures market and compare hedge performance of it with that of CC-GARCH model, VECM model, VAR model and OLS model. Hedging effectiveness is measured using the percentage of variance reduction and the percentage of sharp ratio reduction. We find that in terms of two different measurement of hedging effectiveness, state space model based on kalman filter significantly perform better than other models. The conclusion is robust to hedge periods. The results of the comparison of dynamic CC-GARCH model with static models depend on the duration of the hedge. VECM model perform worst and the hedging performance of VAR model does not significantly surpass that of simple OLS model. The risk of econometric models includes model-misspecification risk and estimation risk. Although the model-misspecification risk of advanced econometric model may be smaller than simple model, its estimation risk is greater and the total risk is uncertain. That the solution of Kalman filter accord with Bayesian rules make Kalman filter approach outperform other models in dealing with estimation risk.The risk of econometric models includes model-misspecification risk and estimation risk. Backward-looking econometric models based on frequentist statistics don't account for the existence of estimation risk. The Bayesian approach provides a general framework where estimation risk is naturally accounted for when considering the parameters as random variable. This fourth chapter use Bayesian approach based on MCMC simulation to estimate the optimal hedge ratio of china's copper futures market. The performance of the Bayesian hedge ratios is compared to that of alternative frequentist statistics approach. The Bayesian empirical result indicate EC-VAR model perform best and the hedging performance of VAR model significantly surpass that of simple OLS model. On the contrary, if not accounting for estimation risk, EC-VAR model perform worst and the hedging performance of VAR model don't significantly surpass that of OLS model The positive correlation between hedging horizon and hedging effectiveness is identified in this research.The optimal hedge ratio depends on the particular objective function to be optimized. In the fifth chapter, we propose four approaches—minimum-variance, maximum-sharp ratio. HKL mean-variance utility function and minimum-VaR—as objective function for hedgers to derive optimal hedge ratio. The bivariate constant correlation GARCH model is applied to estimate the within-sample optimal hedge ratios. The hedging performance of these four hedge ratios is evaluated by portfolio coefficient of variation. The empirical results indicates that the martingale process do not hold for china' copper futures and portfolio variances display a tendency to increase as the risk-averse level decrease. As for hedging performance, the MV hedge ratio does worst and the HKL mean-variances utility function provides better hedging performance than minimum-VaR.Suppose spot price follows Wiener process, the sixth chapter derived the fluctuation process of spot price by using Ito lemma. Based on cost of carry theory, this chapter simulate the coming track of futures price and spot price by the method of Monte Carlo simulation. The empirical results indicates that the sign of the mean of cost of carry in china' copper futures market is negative. The find is consistent of backwardation theory proposed by Keynes. Unfortunately, the coefficient of correlation of the simulated futures return and spot return nearly is zero and the discover make out that china' copper futures market is dominated by speculative trading volume.Objective function of hedge depends on behavioral characteristic of hedger. The behavioral Financial theories find the exist of disappointment aversion of investor. Based on the model of disappointment aversion non-expected utility, the seventh chapter study the effect of disappointment aversion and risk aversion on optimal hedge ratio. The research indicate that disappointment aversion has no effect on the hedging decision when the market is unbiased. In case of backwardation of contago, both disappointment aversion and risk aversion result in more cautious behavior and an optimal contract size closer to the (adjusted) full hedge. Another interesting finding is that a more risk or disappointment aversion hedger will choose a lower reference point.The last chapter is the conclusion and prospect of research.