WSEAS Transactions on Information Science and Applications
Print ISSN: 1790-0832, E-ISSN: 2224-3402
Volume 16, 2019
Comparative Analysis of Classification Algorithms on Stock Market Price Changes
Author:
Abstract: Price change on stock market is very important indicator for investors. In this paper, six Canadian banks’ daily stock market price changes are classified by seven data mining algorithms using Weka. Thirty-seven years of data from 1980 to 2017 obtained from NASDAQ for six Canadian banks with 21 independent variables and one dependent variable, price, are used to classify the daily stock price changes. The numerical data, daily price changes, are converted to nominal data as “up”, “down” and “same” observing the daily price changes according to previous day closing price. To determine which method makes the better classification, all methods run separately for each bank. Then to test the reliability of the techniques, each technique run and compared the original 2018 data. It is seen that, among the seven methods, individually and overall J48 classifies the stock price changes well. Moreover, the results show that J48 algorithm is a promising alternative to the conventional methods for financial prediction
Search Articles
Keywords: Classification, Logistic regression, Fuzzyrough-NN, Genetic Programming, J48, Random Forest, Navie Bayes, Navie Net, Weka, Data mining
Pages: 174-184
WSEAS Transactions on Information Science and Applications, ISSN / E-ISSN: 1790-0832 / 2224-3402, Volume 16, 2019, Art. #20