Visualization Map and Evaluation of Indian Industries Using Statistical Data Mining Techniques

Abstract

Abstract: The uncovering of company and financial difficulties is a theme which has particularly susceptible to financial ratio analysis. Rating and predictive the show of the top ranking companies on the basis of certain financial ratios based on Statistical Data Mining (SDM) techniques. In this paper we used three different Statistical Data Mining methods and they are Factor Analysis, K-means Clustering techniques and Multivariate Disriminant Analaysis. It is well known that statistical information on financial ratios is being extensively used by researchers for many purposes. The financial information of public and private sector companies rated as the best with reference to net sales, it was published by capital market, were considered for the period from 2001 to 2010 for the present study. Out of many possible ratios, eight financial ratios with different notions of the objectives and significant meaning in the literature were selected. To explore the financial data, data mining tools such as factor analysis, k-means clustering and discriminant analyses are applied in succession. Factor analysis is initiated first to uncover the structural models underlying financial ratios. The scores from extracted factors were then used to find initial groups by k-means clustering algorithm to prune the data. The cluster analysis was followed by iterative discriminant procedure with original ratios until cent percent classification was achieved. Finally, the groups were identified as companies belonging to Grade H, Grade M and Grade L in that order, which show the behavior of High performance, Moderate performance and Low performance. Similarly the companies belonging to Grade M category are superior to those of Grade L but inferior to Grade H, indicating the members in the category Grade L are at the low profile in terms of the importance of some of the parameters considered for the present study. The pictorial representations of the groups are also depicted. Industrial analyst can make use of the grading of industries proposed in this paper to know the performance of companies and for the purpose of investment.