ABSTRACT

Classification of Indian stock market data has always been a certain appeal for researchers. In this paper, first time combination of three supervised machine learning algorithms, classification and regression tree (CART) , linear discriminant analysis (LDA) and quadratic discriminant analysis (QDA) are proposed for classification of Indian stock market data, which gives simple interpretation of stock market data in the form of binary tree, linear surface and quadratic surface respectively. These resulted forms help market analyst to make decision on selling, purchasing or holding stock for a particular company in Indian stock market. In section IV and V, experimental results and performance comparison section show that classification and regression tree misclassification rate is only 56.11% whereas LDA and QDA show 74.26% and 76.57% respectively. Smaller misclassification reveals that CART algorithm performs better classification of Indian stock market data as compared to LDA and QDA algorithms.