The performance of implied volatility and historical volatility in forecasting future realized volatility : an analysis from Malaysia, Singapore and Thailand structured call warrants

Abstract

Volatility is a most critical concept in both the theory and practice of finance. Two mainly known volatility estimator are historical volatility or past realized volatility and Implied volatility. Implied volatility is widely seen as the market’s estimates of future volatility, and if markets are efficient, it should thus reflect all the information given at that particular time, including that contained in historical volatility. On the other hand, historical volatility is an unconditional predictor that ignores the most recent publicly available data. Contrasting views from past researches on which volatility estimator is superior in forecasting future realized volatility has prompted this study. The objectives of this study to examine the information content and predictive power of the implied volatility of structured call warrants from three ASEAN countries, namely Malaysia, Singapore and Thailand, in the period of August 2014 to July 2015. Specifically, the aim of the study is to investigate the ability of the structured call warrants explanatory variables to forecast (1) 1 day volatility and (2) volatility over the remaining days of the structured warrants contract. Two empirical models, time series historical and implied volatility, were estimated from the structured call warrants. All samples were then sorted by the volatility, time to maturity and moneyness, thus creating 50 subsamples for each country. Next, these volatilities are assessed through Ordinary Least Squares (OLS) assumptions - this methodology enables the addressing of the informational content, the biasness and efficiency of the forecast predictor. Additionally, an in-sample forecasting accuracy test is employed to identify the most efficient forecasting model. This study found that for Malaysia, Singapore and Thailand, both implied and historical volatility were biased and inefficient predictors of future realized volatility. Implied volatility does not incorporate all the information on future realized volatility, so does historical volatility. However, historical volatility had more predictive power than implied volatility when forecasting future realized volatility. Finally, the in-samples forecast accuracy also showed that the forecasting capabilities were poor for all the three ASEAN countries. This study adds to the growing literature of implied volatility. It also contributes new evidence to the academician and practitioner on forecasting capabilities of implied volatility from the emerging market, especially the ASEAN region. In addition, this study can be the impetus for new research on different model of implied volatility forecasting competing against more sophisticated historical volatility model. Keywords: Implied volatility; historical volatility; options; forecasting