An Evaluation of Volatility Forecasting Techniques

Abstract

The existing literature contains conflicting evidence regarding the relative quality of stock market volatility forecasts. This paper employs daily Indian data to examine the relative ability of various models to forecast monthly stock market volatility. The forecasting models which were selected range from naive model to relatively complex GARCH model. While it is difficult to claim superiority of any one model under all measures used to assess the accuracy of the forecast, the overall results clearly identify two competing models i.e., the RWM and GARCH (1, 1).