ABSTRACT
Data mining of stock prices has a long history. Traders have always looked for patterns and other indicators that they think may help to predict stock-price movements. The term “data mining” here refers to exploratory data analysis. In data mining, the analyst is not just fitting the coefficients in a regression model or estimating
the autoregressive order in a time series. Although formal methods of statistical inference may be part of the process, in data mining the goals are somewhat less focused. The hope is to discover the unexpected. A common characteristic of an exercise in data mining is the use of data from multiple sources, in different formats, and collected for different purposes.
