hrvatski

Evaluation of Alternative Approaches in Classification Algorithms for Prediction of Stock Market Index

This chapter tackles the problem of automatic recognition of favorable days for intra-day trading. The problem is modeled as a binary classification problem and several approaches are tested for solving it. Croatian stock index CROBEX data is used and 22 technical indicators are calculated as predictor variables. Performance of five classifiers is evaluated and compared by using Cohen’s kappa as evaluation metric: artificial neural network, support network machine, random forest, k-nearest neighbors and naïve Bayes classifier. The results give insight to effectiveness of technical analysis in predicting the day favorability for CROBEX index and suggest that technical analysis makes sense and might work for this case.

Day Trading, Classification Problem, Neural Network, Support Vector Machine, Random Forest, K-Nearest Neighbors, Naïve Bayes, Crobex

nije evidentirano