Development and approbation of a software solution for the investment portfolios formation using fractal analysis and predictive models

This paper describes the development and approbation of software solution, which implements methodology forming recommendations for composition investment portfolios using fractal analysis long memory predictive models, is result research conducted over several years at Department Information Systems Mathematical Methods in Economics PSU. The general algorithm program includes four main steps: 1) obtaining preparing data; 2) sorting assets by dimension their price series; 3) forecasting asset returns; 4) (determining shares). described stages correspond to structure expressed set its subsystems. features developed are as follows: possibility load data share prices from FINAM website; calculation series DFA minimum coverage methods; returns ARFIMA ARFIMA-GARCH models; selection portfolio structures based both on forecasted historical multiple generation random-selected with characteristics averaging; evaluation test (if any); support multi-core processors acceleration calculations; windowed graphical interface. results Russian stock market under crisis economic conditions given. On whole, these line those obtained earlier US data. can be used investors, carrying out activities international financial markets. Scientific application important: users scientists, students other researchers innovative methods formation.