Cornish-Fisher Expansion on Estimation and Forecasting Models of Stock Return Volatility

Leila  Y. Calderon,

Cornish-Fisher Expansion on Estimation and Forecasting Models of Stock Return Volatility

Abstract:

The ability to forecast financial market volatility is important for portfolio selection and asset management. However, predicting volatility is a challenge and there are different volatility models available to choose from. Understanding the stock return volatility will result into better investment strategies. In most cases, the widespread popularity of mean-variance analysis is due to the fact that it is very simple and powerful. Most of the theoretical and empirical work on portfolio selection and the pricing of financial assets use the mean-variance analysis. Many investors limit their decisions on the mean-variance analysis as a simple exercise to risk-return tradeoff. The variance is usually unconditional as computed by the standard deviation of the sample period. However, it has been proven by studies that stock return volatility is time varying as was shown by Autoregressive Conditional Heteroscedasticity (ARCH) (Engle, 1982).