High-low Strategy of Portfolio Composition using Evolino RNN Ensembles

Jelena Stankeviciene; Nijole Maknickiene; Algirdas Maknickas

doi:10.5755/j01.ee.28.2.15852

Authors

Jelena Stankeviciene Vilnius Gediminas Technical University Faculty of Business Management Department of Finance Engineering
Nijole Maknickiene Vilnius Gediminas Technical University Faculty of Business Management Department of Finance Engineering
Algirdas Maknickas Vilnius Gediminas Technical University Faculty of Fundamental Sciences Department of Information Technologies

DOI:

https://doi.org/10.5755/j01.ee.28.2.15852

Keywords:

finance markets, Evolino, high-low strategy, investment portfolio, prediction

Abstract

The originally configured 176 Evolino recurrent neural networks (RNN) connected to one ensemble and trained in parallel is an artificial intelligence solution, which allows the successful application of this tool for forecasting financial markets. Predictions made by this tool twice a day with different historical data give two distributions of expected values, which reflect future dynamic exchange rates. Constructing the portfolio, according to shape, parameters of distribution and the current value of the exchange rate allows the optimization of trading in daily exchange-rate fluctuations. Comparison of a high-low portfolio with a close-to-close portfolio shows the efficiency of the new forecasting tool and new proposed trading strategy.

DOI: http://dx.doi.org/10.5755/j01.ee.28.2.15852

Author Biographies

Jelena Stankeviciene, Vilnius Gediminas Technical University Faculty of Business Management Department of Finance Engineering

Stankevičienė Jelena. Vilnius Gediminas Technical University, Saulėtekio a, LT–10223, Vilnius, Lithuania. Faculty of Business and Management, Department of Finance Engineering, professor, doctor of social science. Her main research topics include assets and liability management, regulation of financial institution, financial management for value creation, value engineering.

Nijole Maknickiene, Vilnius Gediminas Technical University Faculty of Business Management Department of Finance Engineering

Maknickienė Nijolė. Vilnius Gediminas Technical University, Saulėtekio a, LT–10223, Vilnius, Lithuania. Faculty of Business and Management, Department of Finance Engineering, assoc. prof. doctor of social science. Research interests – economic forecasting, investment portfolio as an instrument for resource allocation application in various areas of the economy, investment portfolio optimization, artificial intelligence application on forecasting and management of economic processes.

Algirdas Maknickas, Vilnius Gediminas Technical University Faculty of Fundamental Sciences Department of Information Technologies

Maknickas Algirdas. Vilnius Gediminas Technical University, Saulėtekio a, LT–10223, Vilnius, Lithuania. Faculty of Fundamental Sciences, Department of Information Technologies, assoc. prof. doctor of technological sciences. He is expert in parallel computing of dynamic processes. His area of interest covers complexity of computation and artificial intelligence. He has many years of experience in the development of artificial intelligence application software code in finance engineering. He shears his accumulated experience with students by teaching curses of program languages C/C++, C# also human and computer interfaces.