A Novel GA-SVR Time Series Model Based on Selected Indicators Method for Forecasting Stock Price
| Autor: | Huei-Yuan, Shiu, 許惠媛 |
|---|---|
| Rok vydání: | 2014 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 102 Forecasting stock price is always the hottest topic for investors. In recent years, many time series models have widely been used in forecasting stock price for achieving the smallest loss. However, the previous time series models still have some problems as follows: (1) the selected important technical indicators depend on subjective experiences and opinions; (2) conventional statistical models must satisfy assumptions about variables in data analysis; (3) conventional time series models only considered the one variable and linear variable; (4) hard to determine the parameters of SVR. In order to solve these problems mentioned, this study proposed two novel GA-SVR time series models based on selected indicators method for forecasting stock price. The two proposed models of this study adopted stepwise regression and multivariate adaptive regression splines to select the important indicators. Then, constructing the forecasting model by SVR, and using GA to optimize the forecasting model. In order to evaluate the forecasting performance of proposed models, this study selected the leading enterprises in various industries are CHT, China Steel, Cathay Financial Holdings, Hon Hai and TSMC. The stock prices of leading enterprises from 2003 to 2012 years are collected as experimental dataset and used to verify the accuracy of proposed models. Finally, the root mean square error (RMSE) adopted as evaluation criterion to compare with other models. The experimental results show that the proposed methods combines multi-factor selection and adaptive GA-SVR can have higher accuracy than other models. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|