A Review on Applied Data Mining Techniques to Stock Market Prediction

Introduction

Due to the stock market volatility needs to be captured from hidden information in the database, forecasting stock market movement for intelligent decision making is a quite difficult task. One of the most important problems in stock market is finding efficient mining techniques to have useful information in the stock market movements for investment decisions. A great number of stocks are traded in the stock markets. Therefore the amount of valuable data is generated by the market and this valuable data have some specific dynamic features, for instance expanding in every moment.

Data mining deals with extracting meaningful information from the large amount of data. Investors use historic data of assets, in order to decide how much of their capitals should be allocated between the assets. They try to find hidden patterns of the historic data, which provide information about future movements of assets. Forecasting financial time series is a very complex and challenging problem. Therefore the forecasting procedure requires specific methods of data mining (Kovalerchuk & Vityaev, 2005).

Data mining involves looking in the data for such factors as (Chung & Gray, 1999):

• •

  Associations: Things done together.

• •

  Sequences: Events occurring over time.

• •

  Classifications: Rules for recognizing patterns.

• •

  Clusters: Defining new groups.

• •

  Forecasting: Predictions from time series e.g., stock price movements.

In Association, the relationship of a particular item in a data transaction on other items in the same transaction is used to predict patterns. In Classification, the methods are intended for learning different functions, which map each item of the selected data into one of a predefined set of classes. Cluster analysis takes ungrouped data and uses automatic techniques to put this data into groups. Clustering is unsupervised and does not require a learning set. Prediction analysis is related to regression techniques. The key idea of regression analysis is to discover the relationships of an independent variable with the dependent and other independent variables. Sequential Pattern analysis seeks to find similar patterns in data transaction over a business period (Olson & Delen, 2008).

Classification and prediction are two forms of data analysis, which can be used to extract models describing important data classes or to predict future data trends (Han & Kamber, 2006). Practitioners and researches pay attention to analysis and prediction of future values and trends of the financial market. Many studies on stock market prediction using data mining techniques have been performed during the past decade. The stock market data has tremendous noise and complex dimensionality (Kim & Han, 2000). Some data mining applications have been demonstrated on stock markets (Keogh & Kasetty, 2003; Hajizadeh, Ardakani & Shahrabi, 2010; Atsalakis & Valavanis, 2009).