Constructing of Optimal Portfolio on Russian Stock Market Using Nonparametric Method – Classification and Regression Tree

Abstract

Elena A. Bujanova - National Research University "Higher School of Economics", Department of Finance.

E-mail: ebuyanova@hse.ru

Arthur Rakikovich Sarkisov - National Research University "Higher School of Economics", Department of Finance.

Email: a.sarkisov@inbox.ru

Constructing of optimal portfolio is one of the most important subject in the modern finance theory. The main goal in constructing of optimal portfolio is to create model, which will be able to forecast dynamic of the asset’s return. However, the problem of creating such model is still actual. In many works there are models, which perfectly explained past dynamic of the asset’s return, but don’t match to forecast it. One of the reason of that is sometime there are no appropriate data for analysis ( not enough observations or «noisy» data). In this case traditional parametric methods will give biased estimation.In this paper we used nonparametric method ( classification and regression tree) for analyzing  50 stocks, which are included in calculation base of MICEX stock index. Using of this method allowed us to use not only macroeconomic and technical factors, but also factors with limited data set (factors of fundamental analysis). As a result we constructed optimal portfolio with average return on 7% higher than market portfolio with the same risk. It should be mentioned that in this research we don’t use traditional definition of optimal portfolio which implies comparison of constructed portfolio with market portfolio. In general case this comparison is not correct because market portfolio is more diversified and, hence, have less risk. Therefore we use another definition of optimal portfolio: optimal portfolio is the portfolio with certain risk which return is not less than the return of market portfolio with  the same risk.Also in this research we conducted a comparative analysis of the influence of the factors on stock return. As the result we showed that Russian stock market has features of speculative market, because the most important factors for stock return of Russian stocks are momentum, bid – ask spread and oil price.