Forecasting Volatility and Analyzing the Features of Volatility by three different methods- empirical study based on SSE 50ETF
Abstract China financial capital market has attracted great attention since it was booming fast. Until 2016 (after 26 years the Shanghai Stock Exchange Market was established), the value of Shanghai Stock Exchange Market reached 29463744 billion RMB which roughly equals to 3440532 billion pounds...
| Main Author: | |
|---|---|
| Format: | Dissertation (University of Nottingham only) |
| Language: | English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | https://eprints.nottingham.ac.uk/57520/ |
| _version_ | 1848799483821817856 |
|---|---|
| author | Zou, YuanFang |
| author_facet | Zou, YuanFang |
| author_sort | Zou, YuanFang |
| building | Nottingham Research Data Repository |
| collection | Online Access |
| description | Abstract
China financial capital market has attracted great attention since it was booming fast. Until 2016 (after 26 years the Shanghai Stock Exchange Market was established), the value of Shanghai Stock Exchange Market reached 29463744 billion RMB which roughly equals to 3440532 billion pounds under current exchange rate, which ranks fifth in the world. In 9th February 2015, the first option in China financial capital market, which is SSE 50ETF option, has been official listed on the Shanghai Stock Exchange. It means that Chinese financial capital market had developed into more marketization. It is necessary to describe how the volatility will change in the future time point through to analysis the changes of the historic volatility or information of options contracts, since the volatility is considered as the most significant feature to measure the risks of portfolio. Pan and Poteshman (2008) conclude that the volatility can affect the option prices. French, Schwert, and Stambaugh (1987) also examine the relationship between volatility and stock return. Therefore, volatility is worth to forecast the volatility of stock because it could assist the stock investors to against risk and manage their portfolio more efficient.
However, it is hard to forecast the volatility in the stock market, because there are many limitations on different methods. Firstly, volatility is constant, when we use classic Black-Scholes model. However, volatility should change varying time (Dumas, Fleming, and Whaley, 1998). Secondly, different forecasting models are suitable for different terms of volatility. GARCH-type model has better ability of predicting in short term, while implied volatilities are more efficient in the prediction of future volatilities for over 1 month (Koopman, Jungbacker and Hol, 2005). However, stock option volatility includes more information, such as information of option contracts. Pong, Shiuyan, et al (2004) compare three types of models and find that intraday rates have most accurate forecasts in short term, while implied volatility and historical volatility are similar in long term predicting. As the representative of emerging market, the financial capital market in China is unique but imperfect, it is hard to apply some popular forecasting methods to this market without analyzing (Yang, Yang and Zhou, 2012). Thus, this dissertation is trying to analysis which forecasting method of three will provide better performance in China stock market.
This dissertation collects the data of daily price of SSE 50ETF from WIND database. We find the forecasting value of historical volatility (collected from daily return), realized volatility (the sum of squared intraday return) and implied volatility (collected from option contracts). Firstly, we applied GARCH model and EGARCH model to analysis the leptokurtosis and fat-tail, volatility clustering and leverage effect of SSE 50ETF. Because GARCH model is consider as the best way to describe the feature of volatility such as leptokurtosis and fat-tail, volatility clustering, besides that EGARCH model can also test the asymmetry of series. Secondly, we considered long-memory by HAR-RV model for realized volatility, which is regarded as most accurate volatility estimation. Thirdly, we used Black-Scholes model to compute implied volatility and observed volatility smile and term structure of the SSE 50ETF.
Compared with three different types of volatility, we found that the features of these volatilities, such as leptokurtosis and fat-tail, volatility clustering, leverage effect and long memory, in the China market. And these three methods both have good ability to predict volatility.
Key words: forecasting volatility of options, GARCH-family model, HAR-RV model, Implied Volatility of options |
| first_indexed | 2025-11-14T20:36:23Z |
| format | Dissertation (University of Nottingham only) |
| id | nottingham-57520 |
| institution | University of Nottingham Malaysia Campus |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-14T20:36:23Z |
| publishDate | 2019 |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | nottingham-575202022-11-30T12:33:44Z https://eprints.nottingham.ac.uk/57520/ Forecasting Volatility and Analyzing the Features of Volatility by three different methods- empirical study based on SSE 50ETF Zou, YuanFang Abstract China financial capital market has attracted great attention since it was booming fast. Until 2016 (after 26 years the Shanghai Stock Exchange Market was established), the value of Shanghai Stock Exchange Market reached 29463744 billion RMB which roughly equals to 3440532 billion pounds under current exchange rate, which ranks fifth in the world. In 9th February 2015, the first option in China financial capital market, which is SSE 50ETF option, has been official listed on the Shanghai Stock Exchange. It means that Chinese financial capital market had developed into more marketization. It is necessary to describe how the volatility will change in the future time point through to analysis the changes of the historic volatility or information of options contracts, since the volatility is considered as the most significant feature to measure the risks of portfolio. Pan and Poteshman (2008) conclude that the volatility can affect the option prices. French, Schwert, and Stambaugh (1987) also examine the relationship between volatility and stock return. Therefore, volatility is worth to forecast the volatility of stock because it could assist the stock investors to against risk and manage their portfolio more efficient. However, it is hard to forecast the volatility in the stock market, because there are many limitations on different methods. Firstly, volatility is constant, when we use classic Black-Scholes model. However, volatility should change varying time (Dumas, Fleming, and Whaley, 1998). Secondly, different forecasting models are suitable for different terms of volatility. GARCH-type model has better ability of predicting in short term, while implied volatilities are more efficient in the prediction of future volatilities for over 1 month (Koopman, Jungbacker and Hol, 2005). However, stock option volatility includes more information, such as information of option contracts. Pong, Shiuyan, et al (2004) compare three types of models and find that intraday rates have most accurate forecasts in short term, while implied volatility and historical volatility are similar in long term predicting. As the representative of emerging market, the financial capital market in China is unique but imperfect, it is hard to apply some popular forecasting methods to this market without analyzing (Yang, Yang and Zhou, 2012). Thus, this dissertation is trying to analysis which forecasting method of three will provide better performance in China stock market. This dissertation collects the data of daily price of SSE 50ETF from WIND database. We find the forecasting value of historical volatility (collected from daily return), realized volatility (the sum of squared intraday return) and implied volatility (collected from option contracts). Firstly, we applied GARCH model and EGARCH model to analysis the leptokurtosis and fat-tail, volatility clustering and leverage effect of SSE 50ETF. Because GARCH model is consider as the best way to describe the feature of volatility such as leptokurtosis and fat-tail, volatility clustering, besides that EGARCH model can also test the asymmetry of series. Secondly, we considered long-memory by HAR-RV model for realized volatility, which is regarded as most accurate volatility estimation. Thirdly, we used Black-Scholes model to compute implied volatility and observed volatility smile and term structure of the SSE 50ETF. Compared with three different types of volatility, we found that the features of these volatilities, such as leptokurtosis and fat-tail, volatility clustering, leverage effect and long memory, in the China market. And these three methods both have good ability to predict volatility. Key words: forecasting volatility of options, GARCH-family model, HAR-RV model, Implied Volatility of options 2019-12-01 Dissertation (University of Nottingham only) NonPeerReviewed application/pdf en https://eprints.nottingham.ac.uk/57520/1/ID%EF%BC%9A4316170%EF%BC%9BModule%20Code%EF%BC%9AN14031%EF%BC%9BForecasting%20Volatility%20and%20Analyzing%20the%20Features%20of%20Volatility%20by%20three%20different%20methods-%20empirical%20study%20based%20on%20SSE%2050ETF.pdf Zou, YuanFang (2019) Forecasting Volatility and Analyzing the Features of Volatility by three different methods- empirical study based on SSE 50ETF. [Dissertation (University of Nottingham only)] Forecasting volatility of options GARCH-family model HAR-RV model Implied Volatility of options |
| spellingShingle | Forecasting volatility of options GARCH-family model HAR-RV model Implied Volatility of options Zou, YuanFang Forecasting Volatility and Analyzing the Features of Volatility by three different methods- empirical study based on SSE 50ETF |
| title | Forecasting Volatility and Analyzing the Features of Volatility by three different methods- empirical study based on SSE 50ETF |
| title_full | Forecasting Volatility and Analyzing the Features of Volatility by three different methods- empirical study based on SSE 50ETF |
| title_fullStr | Forecasting Volatility and Analyzing the Features of Volatility by three different methods- empirical study based on SSE 50ETF |
| title_full_unstemmed | Forecasting Volatility and Analyzing the Features of Volatility by three different methods- empirical study based on SSE 50ETF |
| title_short | Forecasting Volatility and Analyzing the Features of Volatility by three different methods- empirical study based on SSE 50ETF |
| title_sort | forecasting volatility and analyzing the features of volatility by three different methods- empirical study based on sse 50etf |
| topic | Forecasting volatility of options GARCH-family model HAR-RV model Implied Volatility of options |
| url | https://eprints.nottingham.ac.uk/57520/ |