This paper examines how volatility responds to return news in the context of stochastic volatility (SV) using a nonparametric method. The correlation structure in the classical leverage SV model is generalized based on a linear spline. In the new model the correlation between the return innovation and volatility innovation is dependent on the type of news arrived to the market. Theoretical properties of the proposed model are examined. A simulation-based maximum likelihood method is developed to estimate the new model. Simulations show that the estimation method provides reliable parameter estimates. The new model is fitted to daily and weekly data in the US and compared with the classical SV models in terms of their in-sample and out-of-sample performances. Empirical results suggest strong evidence in favor of the proposed model. In particular, the new model finds strong evidence of leverage effect when the classical model fails to identify it. Also, the new model provides better out-of-the-sample forecasts of volatility than the classical model.
Leverage effect, Simulated maximum likelihood, Laplace approximation, Spline, Realized volatility
Econometrics | Statistics and Probability
A Semiparametric Stochastic Volatility Model. (2008). 1-27. Research Collection School Of Economics.
Available at: http://ink.library.smu.edu.sg/soe_research/1268
Copyright Owner and License
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.