Applying the GWMA Method in Forecasting the Stock and Index Volatility
碩士 === 國立中興大學 === 統計學研究所 === 106 === The application of Exponentially Weighted Moving Average (EWMA) is said to have more accurate model predictions which minimizing the variations in the chronological order. More and more papers reveal that the Generally Weighted Moving Average (GWMA) is more sensi...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2018
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/ya4628 |
| Summary: | 碩士 === 國立中興大學 === 統計學研究所 === 106 === The application of Exponentially Weighted Moving Average (EWMA) is said to have more accurate model predictions which minimizing the variations in the chronological order. More and more papers reveal that the Generally Weighted Moving Average (GWMA) is more sensitive with micro measurement because of the parameter α, and can be applied in vast fields. The purpose of this paper is to find out the best parameter value within EWMA and GWMA which will be reviewing the predicted volatilities and the two prediction models were compared with the standard value is based on the historical volatility and AR(1)-GARCH(1, 1).
The US stock market has always been considered as a leading indicator and has a significant impact on Taiwanese stocks. The data of this paper was the volatility between US index and Taiwan index and the volatility of Taiwan index and Taiwan stock. This study exacted the top three constituent stock - Taiwan Semiconductor Manufacturing Corporation (TSM), Hon Hai and South Asia Plastics Industry (NPC). It can be concluded that the effectiveness of GWMA model has better prediction results than the EWMA model.
|
|---|