Data Mining Applying In Stock Trade Analysis

The research works of this paper mainly focus on how to apply data warehouse/data mining technologies in stock market analysis. The main contents of this paper include:1. Designed a flexible and wieldy software application solution for stock market analysis based on data warehouse/data mining technologies. With this solution, can effectively combine the traditional stock technical analysis method and the data warehouse/data mining technologies, and can conveniently analysis, verify all kinds of technical indicators and compound them to predict the market trend.2. Designed and developed an entire stock analysis data warehouse logical and physical data models and corresponding ETL program. These models can effectively support the current analysis requirement. With good expanding ability, they are easily to be enriched with new data and new method.3. Proposed and implemented a new algorithm which can define, identify and present the relative high and low points of stock prices. This algorithm can describe the trend changing of stock prices precisely, and this is a crucial key point for modeling many kinds of market analysis method.4. Verified the effectiveness of widely used MACD, KDJ, RSI indicators. Confirmed their validity, and also found some mistakes of traditional analysis method.5. Based on the associated rules algorithm and neural network algorithm of Data mining technologies, created mining models which can predict the trends changing of stock prices. Verified the effectiveness of these models in reality condition.All the applications in this paper are based on Microsoft SQL SERVER 2005 and Visual Studio .NET platform.