Using Association Rule Mining to Study the Return Relationships Among Stocks
碩士 === 國立臺北科技大學 === 資訊與財金管理系碩士班 === 104 === A variety of tools and methods were used to predict the changes in stock prices. Choosing an appropriate method of analysis is the key to getting profit. In recent years, finding the correlation of stock markets to provide significant rules for the financi...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Online Access: | http://ndltd.ncl.edu.tw/handle/3g8w4t |
| Summary: | 碩士 === 國立臺北科技大學 === 資訊與財金管理系碩士班 === 104 === A variety of tools and methods were used to predict the changes in stock prices. Choosing an appropriate method of analysis is the key to getting profit. In recent years, finding the correlation of stock markets to provide significant rules for the financial market is an important direction of research.
However, there are many ways of predicting stock market. Most of them were studying the price relationships among stocks. The study is using association rule mining to study the return relationships among stocks by using R software. And the study add Rule Interestingness to find valuable rules. To provide more specific information, the rules compared with the actual stock data and be verified at the same time. The Source1 in this study included 118 trading days and generated 223 rules; the Source2 in this study included 991 trading days and generated only 12 rules. The results show that the rules in different data sources diverge greatly. The returns between companies have relationships to some extent.
|
|---|