TITLE
Algebraic models for time series prediction based on evolutionary algorithms
THE SCOPE OF THE RESEARCH
The application of algebraic models for prediction of real world time series is a difficult problem due to the
sensitivity of these models to the additive noise. It is well known that algebraic interpolants are not
applicable for practical extrapolation due to various numerical effects (Runge phenomenon for example).
Intelligent conciliation of algebraic models with smoothing prediction models (moving averaging method for
example) enables to find a compromise between the roughness of prediction produced by algebraic
techniques and smoothness of MA. However, a direct merging of such different techniques is possible only
by using evolutionary algorithms for the construction of adaptive algebraic skeletons.
The expected outcome is the development of new time series forecasting models for more accurate and
efficient prediction of real world time series.
SUPERVISOR
Dr. Habil. Minvydas Ragulskis, Department of Mathematical Modelling, Faculty of Mathematics and
Natural Sciences, Kaunas University of Technology
SUPERVISOR’s COMPETENCE
Papers (2012-2016)
1. M. Landauskas, M. Ragulskis. Clocking convergence to a stable limit cycle of a periodically driven
nonlinear pendulum. Chaos: An Interdisciplinary Journal of Nonlinear Science. 2012, vol. 22, article
no. 033138.
2. Z. Navickas, R. Smidtaite, A. Vainoras, M. Ragulskis. The explosive divergence in iterative maps of
matrices. Communications in Nonlinear Science and Numerical Simulation. 2012, vol. 17, Issue 11,
2012, pp. 4430 - 4438.
3. V. Petrauskiene, A. Aleksa, A. Fedaravicius, M. Ragulskis. Dynamic visual cryptography for optical
control of vibration generation equipment. Optics and Lasers in Engineering. 2012, vol.50, Issue 6,
2012, pp.869-876.
4. V. Petrauskiene, A. Aleksa, A. Fedaravicius, M. Ragulskis. Dynamic visual cryptography for optical
control of vibration generation equipment. Optics and Lasers in Engineering. 2012, vol.50, Issue 6,
2012, pp.869-876.
5. Z. Navickas, M. Ragulskis. Comments on "Two exact solutions to the general relativistic Binet's
equation". Astrophysics Space Sciences. 2013, vol.334, pp.281-285.
6. Z. Navickas, L. Bikulciene, M. Rahula, M. Ragulskis. Algebraic operator method for the
construction of solitary solutions to nonlinear differential equations. Communications in Nonlinear
Science and Numerical Simulation. 2013, vol.18, issue 6, pp.1374-1389.
7. R. Palivonaite, K. Lukoseviciute, M. Ragulskis. Algebraic segmentation of short nonstationary time
series based on evolutionary prediction algorithms. Neurocomputing. 2013, vol. 121, p.354-364.
8. A. Palevicius, M. Ragulskis, G. Janusas. High frequency excitation for thermal imprint of
microstructures into a polymer. Experimental techniques. 2013, vol. 37, issue 5, p.41-47.
9. K. Malinauskas, P. Palevicius, M. Ragulskis, V. Ostasevicius. Validation of noninvasive MOEMSassisted measurement system based on CCD sensor for radial pulse analysis. Sensors. 2013, vol.13,
pp.5368-5380.
10. M. Landauskas, M. Ragulskis. A pseudo-stable structure in a completely invertible bouncer system.
Nonlinear Dynamics. 2014, vol. 78, issue 3, p. 1629-1643.
11. Z. Navickas, M. Ragulskis. Comments on "A new algorithm for automatic computation of solitary
wave solutions to nonlinear partial differential equations based on the Exp-function method".
Applied Mathematics and Computations. 2014, vol. 243, p. 419-425.
12. P. Ziaukas, T. Ragulskis, M. Ragulskis. Communication scheme based on evolutionary spatial 2x2
games. Physica A: Statistical Mechanics and its Applications. 2014, vol. 403, p. 177-188.
13. V. Petrauskiene, R. Palivonaite, A. Aleksa, M. Ragulskis. Dynamic visual cryptography based on
chaotic oscillations. Communications in Nonlinear Science and Numerical Simulation. 2014, vol.19,
issue 1, p.112-120.
14. R. Palivonaite, M. Ragulskis. Short-term time series algebraic forecasting with internal smoothing.
Neurocomputing. 2014, vol. 127, p.161-171.
15. V. Petrauskiene, A.Survila, A.Fedaravicius, M.Ragulskis. Dynamic visual cryptography for optical
control of chaotic oscillations. Optics & Laser Technology. 2014, vol. 57, p.129-135.
16. R. Palivonaite, A. Aleksa, A. Paunksnis, A. Gelzinis and M. Ragulskis. Image hiding in timeaveraged deformable moire gratings. Journal of Optics. 2014, vol.16, article no. 025401.
17. P. Palevicius, M. Ragulskis, A. Palevicius, V. Ostasevicius. Applicability of time-averaged
holography for micro-electro-mechanical system performing non-linear oscillations. Sensors. 2014,
vol. 14(1), pp. 1805-1821.
18. M. Vaidelys, J. Ragulskiene, P. Ziaukas, M. Ragulskis. Image hiding scheme based on the atrial
fibrillation model. Applied Sciences. 2015, vol. 5, pp. 1980-1991.
19. D. Karaliene, Z. Navickas, R. Ciegis, M.Ragulskis. An extended Prony’s interpolation scheme on an
equispaced grid. Open Mathematics. 2015, vol.13, p.333-347.
20. P. Palevicius, M.Ragulskis. Image communication scheme based on dynamic visual cryptography
and computer generated holography. Optics Communications. 2015, vol. 335, p. 161-167.
21. M. Vaidelys, J. Ragulskiene, S. Aleksiene, M. Ragulskis. Image hiding in time-averaged moiré
gratings on finite element grids. Applied Mathematical Modelling. 2015, vol. 39, p. 5783-5790.
22. Z. Navickas, T. Telksnys, M. Ragulskis. Comments on "The exp-function method and generalized
solitary solutions." Computers and Mathematics With Applications. 2015, vol. 69, p. 798-803.
23. M. Vaidelys, P. Ziaukas, M. Ragulskis. Competitively coupled maps for hiding secret visual
information. Physica A: Statistical Mechanics and its Applications. 2016, vol. 443, p. 91-97.
24. R. Palivonaite, K. Lukoseviciute, M. Ragulskis. Short-term time series algebraic forecasting with
mixed smoothing. Neurocomputing. 2016, vol. 171, p. 854-865.
25. A. P. Fedaravicius, Maosen Cao, M. Ragulskis. Control of a dendritic neuron driven by a phaseindependent stimulation. Chaos, Solitons & Fractals. 2016, accepted article in press.
26. M. Landauskas, Z. Navickas, A. Vainoras and M. Ragulskis. Weighted moving averaging revisited an algebraic approach. Computational and Applied Mathematics. 2016, accepted article in press.
Projects (2012-2016):
Project Code: MIP-041/2011
Project Title: H-rank for Investigation of Nonlinear Systems – The Concept and Applications
Project Leader: Minvydas Ragulskis
Budget: 41.507,- EUR
Funding Body: Lithuanian Research Council
Duration: 2011.03.01 - 2012.12.31
Description: Breakthrough in the area of the mathematical description of nonlinear systems based on the
concept of H-rank.
Project Code: MIP-100/2012
Project Title: Dynamic Visual Cryptography for the Assessment of Human Visual System
Project Leader: Minvydas Ragulskis
Budget: 52.405,- EUR
Funding Body: Lithuanian Research Council
Duration: 2012.09.03 - 2014.08.30
Description: Development of a PC-based workplace at the Eye Clinics (Kaunas Medical Academy) for the
assessment of Human Visual System based on Dynamic Visual Cryptography.
Project Code: VP1-3.1-SMM-10-V-02-005
Project Title: The Development and Implementation of Training Rocket Equipment
Project Leader: Algimantas Fedaravicius
Role: Minvydas Ragulskis – Senior Research Fellow
Budget: 714.583,- EUR
Funding Body: European Social Fund
Duration: 2013 - 2015
Description: Confidential.
Project Code: MIP-078/2015
Project Title: Algebraic Techniques and Algorithms for Time Series Analysis
Project Leader: Minvydas Ragulskis
Budget: 91.084,- EUR
Funding Body: Lithuanian Research Council
Duration: 2015.04.01 - 2018.03.31
Description: This project is dedicated for the development of new methods and algorithms for time series
analysis and application of these algorithms for the identification, forecasting and segmentation of chaotic
processes.
PROVIDED FUNDING PER YEAR
Postdoctoral Fellow’s salary – 20000 EUR (all taxes included)
Grant for dissemination activities – 4500 EUR