The Selection of Winning Stocks Using Principal Component Analysis

Abstract

One of the primary challenges with stock selection is the identification of the best stock features to use for the selection of winning stocks. Typically, there are easily more than 50 variables that can be used for stock selection. Many stock investors prefer to keep stock selection as simple as possible and therefore are interested in identifying a few stock variables to use for the identification of winning stocks. Principal Component Analysis is a statistical technique that reduces a large number of inputs of data to a few factors. Once the factors are established, they are displayed in a perceptual map. The perceptual map provides a clear picture of the winning stocks that should be selected for trading.