DOI: 10.5176/2251-1997_AF16.50
Authors: Alireza Heidarzadeh Hanzaei
Abstract:
Regarding growing development of novel financial instruments, there has been an increasing call for the theoretical and empirical knowledge of financial volatility. Volatility is not directly observable and must be estimated. But before forecasting the future volatilities, we have to measure what has happened in the past. Measuring is an art and we should find the most suitable method among several historical volatility measurement methods. As we know, sample volume is very important factor for the improvement of estimators forecasting from volatilities. It means that using intraday data; we can estimate volatility in a shorter period of time more precisely using intraday data. Using Intra-Daily data in Tehran Securities Exchange (TSE) between 2011/Sep/21 and 2013/Sep/21, we have studied and chosen the mot efficient estimator among different volatility estimators (Parkinson, Rogers-Satchel, Garman-Klass and Yang-Zhang). The data for this study has been collected in a thirty minutes period between 9 AM to 12 AM and there have been 6 observations (data) for each trading day. In calculating the suitable estimator, considering the (interval used in calculations) and n (efficient time), it has been shown that the suitable selection of these two factors, play an important role in measuring the volatilities. This study results, showed that Parkinson estimator has a better performance than any other estimator using the 30 minutes data and having the data of the day before. Our findings confirmed that the best estimator should include information contained not only in closing prices but in the price range as well (range estimators).
Keywords: Volatility Estimators; Historical Volatility; Relative Volatility Efficiency Ratio; Intra-Day Data
Updating...