AI SearchA decision support system for forecasting of stock prices
Shieh, Hong-zong
A decision support system for forecasting of stock prices - Bangkok : Asian Institute of Technology, 1994 - 73 leaves - Thesis ; no. CS-94-24 . - Asian Institute of Technology. Thesis ; no. CS-94-24 .
A thesis submitted in partial fulfillment of the requirement for the degree of Master of Science, School of Engineering and Technology
Thesis (M.Sc.) - Asian Institute of Technology, 1994
This study presents a stock decision support system (SDSS) developed for forecasting stock prices in Taiwan market. The SDSS improves the decision making process by placing information in the hands of the user at the proper time and to supports a complete flexibility in the choice and sequence of analysis, and in the presentation of the results. The hybrid technique combines the back propagation neural network, Box-Jenkins ARIMA time series method and stepwise regression analysis method to provide the user with a flexible modeling and forecasting environment. The results show in time series forecasting no matter what the lead time equal to one or more the present value is the most important factor of forecasting values. The system assists the user in making the correct and speedy decision in choice of the best buy and sell time of a stock. It has been shown by several simulation results that SDSS system is quite helpful for making a good forecast of stock price.
Stock price forecasting
Decision support systems
A decision support system for forecasting of stock prices - Bangkok : Asian Institute of Technology, 1994 - 73 leaves - Thesis ; no. CS-94-24 . - Asian Institute of Technology. Thesis ; no. CS-94-24 .
A thesis submitted in partial fulfillment of the requirement for the degree of Master of Science, School of Engineering and Technology
Thesis (M.Sc.) - Asian Institute of Technology, 1994
This study presents a stock decision support system (SDSS) developed for forecasting stock prices in Taiwan market. The SDSS improves the decision making process by placing information in the hands of the user at the proper time and to supports a complete flexibility in the choice and sequence of analysis, and in the presentation of the results. The hybrid technique combines the back propagation neural network, Box-Jenkins ARIMA time series method and stepwise regression analysis method to provide the user with a flexible modeling and forecasting environment. The results show in time series forecasting no matter what the lead time equal to one or more the present value is the most important factor of forecasting values. The system assists the user in making the correct and speedy decision in choice of the best buy and sell time of a stock. It has been shown by several simulation results that SDSS system is quite helpful for making a good forecast of stock price.
Stock price forecasting
Decision support systems