Open AccessMulti-factor Stock Selection Model Based on Adaboost
Author(s) -
Ru Zhang,
Tun Cao
Publication year - 2018
Publication title -
business and economic research
Language(s) - English
Resource type - Journals
ISSN - 2162-4860
DOI - 10.5296/ber.v8i4.13942
Subject(s) - adaboost , profitability index , stock (firearms) , artificial intelligence , computer science , classifier (uml) , machine learning , model selection , research object , econometrics , pattern recognition (psychology) , mathematics , economics , engineering , business , finance , mechanical engineering , business administration
In this paper, we established multi-factor stock selection model based on Adaboost by using Adaboost to integrate the custom week classifier model, and Shanghai and Shenzhen 300 stocks are taken as the research object. During the stock retest, the first is make a comparative test between Adaboost multi-factor stock selection model and the traditional multi-factor model, among them, the factor large class isn’t considered in the multi-factor stock selection model. And the results of two contrast experiment showed that the multi-factor stock selection model based on Adaboost has stronger profitability and less risk than the traditional multi-factor model.