The Stock Selection with Ward, K-Means and Two-Steps Clustering Analysis Methods Based on the Financial Indicators

In this study, a data mining approach for classification of stocks into clusters is presented. After classification, the stocks could be selected from these groups for building a portfolio. This study aims to create an effective portfolio from stocks traded in Stock Exchange Istanbul using three different clustering analysis methods. Another purpose of the study is to test the availability of clustering analysis methods to create an efficient portfolio of stocks. For these purposes, a total of 69 stocks were clustered by using Ward method as a hierarchical clustering method, K-Means method as a non-hierarchical clustering method, and two-step clustering (hybrid) method. The financial indicators that used in this study were obtained from financial statements and stock price movements of companies. As a result of the study, clusters formed according to all three methods are generally similar. The clusters are evaluated based on the average of the financial indicators and the number of shares, and the preferable clusters are indicated