The Data Mining Of Finance - 2031 Words

Data Mining in Finance 1. Introduction Data mining is used to uncover hidden knowledge and patterns from a large amount of data. In finance, there is enormous data which generates during business operations and trading activities. Extracting valuable data from them manually might be unable or spend a lot of time. As a result, data mining plays an importance role of discovering data that can be used to increase profits and evaluate risks of strategic planning and investment. The data mining methods in finance are derived from statistics, machine learning, and visualization. The most commonly used methods are Decision Trees, Neural Networks, Genetic Algorithms, and Rough Set Analysis (Hajizadeh, et al., 2010). Due to prediction and†¦show more content†¦Prior forecasting method is the prediction based on the growth rate of fundamental factors such as earning-per-share, book value, and invested capital. Another common method for forecasting is time series analysis. The traditional analysis uses regression models to d iscover market trends. However, financial time series are complex, noisy, and nonlinear pattern. As a result, later techniques have applied artificial intelligences; the most popular technique is neural networks (Zhang Zhou, 2004). According to Refenes’s experiment (1994), the back-propagation networks, a type of neural network, predicts more accurately than regression models on the same datasets. Other techniques are ARMA and ARIMA model, genetic algorithm, and random walk. Pai and Lin (2005) introduces the A hybrid model ARIMA and support vector machine which give better result than single ARIMA. Edward Tsang et al. (1998) proposes EDDIE, a tool which constructs a genetic decision tree to evaluate whether a stock price increases in a specified time. Jar-Long and Shu-Hui (2006) proposes a two-layer bias decision tree with simple technical indicators. Typically, the input variables of prediction models are daily price, volume, rate of change, and technical indicators, for example, moving average (MA), relative strength index (RSI), and volatility (Vol). 2.2 Foreign exchange market Foreign exchange market opens 24 hours a day, 7 days a week. It is the