Modified Exponentially Weighted Moving Average Model to Estimate Stock Volatility in Indonesia Stock Exchange (Case Study: LQ45 Index).

Abstract

Accurate volatility modeling is an important issue in finance. A common approach to estimate the volatility of asset returns is to use an exponentially weighted moving average (EWMA) model, a special case of general autoregressive conditional heteroscedasticity (GARCH) model with optimized smoothing weights. The standard EWMA estimator is based on the maximum likelihood estimator of the variance of normal distribution, and is thus optimal when the returns are conditionally normal. Financial asset returns are well-known to be non-normal and leptokurtic. In this research, we propose an alternative EWMA estimator that is robust to leptokurtosis in the conditional distribution of returns. The estimator is based on the maximum likelihood estimator of variance of the power exponential distribution. It is a function of an EWMA of the absolute value of past returns, rather than their squares. The aim of this paper is to compare forecasting performance of this newly developed estimator named Power EWMA with EWMA and GARCH. The data used in this research are daily closing stock prices of LQ45 index in the period of 2002-2011. We found that the Power EWMA has the best performance of the three models with respect to root mean square forecast error.