Constructing a Stock Trading Decision System Using Bootstrap Methods and Group Methods of Data Handling

碩士 === 國立交通大學 === 工業工程與管理系所 === 106 === Due to the current economic downturn and low wages, the housing prices and commodity prices are high and the bank interest rates are generally below 1%. Therefore, more people are more interested in participating in the stock market than before. There are many...

Full description

Bibliographic Details
Main Authors: Lu, Teh-Wei, 呂得瑋
Other Authors: Tong, Lee-Ying
Format: Others
Language:zh-TW
Published: 2018
Online Access:http://ndltd.ncl.edu.tw/handle/f3t568
Description
Summary:碩士 === 國立交通大學 === 工業工程與管理系所 === 106 === Due to the current economic downturn and low wages, the housing prices and commodity prices are high and the bank interest rates are generally below 1%. Therefore, more people are more interested in participating in the stock market than before. There are many ways of investigating the stock price, including fundamental analysis, technological analysis, etc. Unfortunately, these analytical methods can only be utilized by the investment consulting corporations but not by individual investor who has only limited information about stocks. Therefore, the objective of this study is to develop a system of share transaction for the long-term individual investors in stocks. This study first utilizes some financial indicators, such as capital amount, return on equity, gross profit margin and P/E ratio, to select a group stocks with long-term investment value. The Bootstrap confidence intervals are then constructed for determining the optimal trading price based on the historical P/E ratio of stocks. The Group Method of Data Handling (GMDH) is also employed to predict the fluctuation of stock prices of the next day. Finally, the information obtained by the Bootstrap confidence intervals and the GMDH prediction method is integrated to construct a stock-trading decision system. The individual stock investor uses the proposed stock-trading decision system may trade the stocks easily and gain reasonable profit. Finally, this study uses thirteen Taiwanese stocks to verify the proposed method is indeed effective. 【keyword】Stock Price Forecast, Bootstrap Methods, Group Method of Data Handling, Stock Trading Price, Decision System