Methods for positioning and management of a portfolio of securities

This article explores the problems of forming an investment portfolio using the current classical methodology and suggests its expansion using modern models of accounting for the general mood of the market and unobservable market data. There are two components to consider when building a portfolio: the price dynamics of available investment opportunities and the decision model that controls the return on investment. Average variance analysis, developed by Markowitz in 1952, has been adopted as a paradigm for portfolio selection. Investment risk is measured by the variance of the portfolio return in accordance with the assumption of the normality of asset returns. The normality assumption not only made portfolio investments convenient, but also provided a strong economic theory on which the famous Capital Asset Pricing Model proposed in 1964 by Sharpe was based. However, this medium variance model typically results in large portfolio turnover because the underlying data generation process is far from normal, which limits its applicability to dynamic investment strategies. Mean variance analysis also performs poorly in out-of-sample tests, which confirms that asset returns can be asymmetric, and a different measure of uncertainty is required to characterize asset returns over time, which determined the relevance of the study. The study suggests the introduction of a hidden Markov model inside the classical CAPM model for pricing stratification in different market regimes. The result of this study is the proposal of a new policy, as well as an assessment of the accuracy of the proposed methodology.