Abstract
The aim of the study was to monitor the system theoretic exogenous variables augmented state space algorithm of Aoki (State space modelling of time series. Springer, Heidelberg, 1987) and the VARMAX algorithm of Spliid (J Am Stat Assoc 78(384):843–849, 1983) within a geno-mathematical framework towards optimal parametric conditions/search intervals. Both algorithms were implemented as an integrated support library for a general computational platform, the Genetic Hybrid Algorithm (GHA), where some key parameters of the algorithms are defined in a search process utilizing a mixed geno-mathematical search technique. The empirical results of our tests using real economic data from the European stock market are encouraging. Specifically, the information criteria used in the VARMAX-search (Vector Autoregressive Moving Average algorithm with Exogenous variables) algorithm tend to favor parsimonious model representations automatically. Furthermore, the state space algorithm captures almost the same dynamics as the complex VARMAX-model estimated in the study. Both algorithms have encouraging in sample properties. When generating k-steps forecasts out-of-sample, k > 1, the state space algorithm seems to deteriorate faster than the VARMAX algorithm, however. The results suggest that more empirical testing is needed, especially in different situations with different degrees of model order and stationarity conditions, in order to provide more evidence on the suitability of the competing methods in particular cases. We demonstrated that the Genetic Hybrid Algorithm can be used as a generic platform for parametric search in vector valued time series modelling. Efficient procedures for optimal grouping of the individual time series processes and recognition of heteroskedasticity may improve the performance of the algorithms further.
Similar content being viewed by others
Notes
The Eigen-mass limit 0.01 means that all eigenvalues with proportion \( \frac{{\mathop \lambda \nolimits_{i} }}{{\sum\nolimits_{j = 1}^{h} {\mathop \lambda \nolimits_{j} } }} > 0.01\) are considered to be large.
References
Aoki M (1987) State space modelling of time series. Springer, Heidelberg
Aoki M (1988) State space models for vector-valued time series with random walk components. In: Research paper presented at the eighth international symposium of forecasting. Amsterdam, June
Aoki M, Havenner A (1991) State space modelling of multiple time series. Econom Rev 10(1)
Bollerslev T (1986) Generalized autoregressive conditional heteroskedasticity. J Econom 3:307–327
Box GEP, Jenkins GM (1976) Time series analysis, forecasting and control. Holden-Day
Deistler M (1991) Comment on: state space modelling of multiple time series. In: Aoki M, Havenner A. Econom Rev 10(1):61–65. doi:10.1080/07474939108800195
Dickey D, Fuller W (1979) Distribution of the estimators for autoregressive time series with a unit root. J Am Stat Assoc 74:427–431. doi:10.2307/2286348
Dorfman JH (1991) Comment on: a model specification test for state space models. Econom Rev 10(1):67–73. doi:10.1080/07474939108800196
Edlund P-O (1984) Identification of the multi-input Box-Jenkins transfer functions model. J Forecast 3:297–308. doi:10.1002/for.3980030307
Edlund P-O (1989) Preliminary estimation of transfer function weights: a two-step regression approach. Dissertation, Stockholm School of Economics
Engle R (1982) Autoregressive conditional heteroskedasticity with estimates of the variance of United Kingdom inflation. Econometrica 50(4):987–1007. doi:10.2307/1912773
Ferentinos KP (2005) Biological engineering applications of feedforward neural networks designed and parameterized by genetic algorithms. Neural Netw 18:934–950. doi:10.1016/j.neunet.2005.03.010
Höglund R, Östermark R (1991) Automatic ARIMA modelling by the Cartesian search algorithm. J Forecast 10:465–476. doi:10.1002/for.3980100503
Jenkins GM (1979) Practical experiences with modelling and forecasting time series. G. Jenkins & Partners Ltd, St Helier
Ljung GM, Box GEP (1978) On a measure of lack of fit in time series models. Biometrika 65(2):297–303
Mittnik S (1991) Comment on: state space modelling of time series. Econom Rev 10(1):1–59. doi:10.1080/07474939108800197
Monsell BC (2002) An update on the development of the X-12-ARIMA Seasonal Adjustment Program. In: Proceedings of the third international symposium on frontiers of time series modelling, pp 1–11
Östermark R (1990) Portfolio efficiency of univariate time series models. OMEGA Int J Manage Sci 18:159–169. doi:10.1016/0305-0483(90)90063-F
Östermark R (1999a) Solving irregular econometric and mathematical optimization problems with a genetic hybrid algorithm. Comput Econ 13(2):103–115. doi:10.1023/A:1008621308348
Östermark R (1999b) A Neuro-genetic algorithm for heteroskedastic time series processes. Soft Comput 3(4):206–220. doi:10.1007/s005000050010
Östermark R (2002) A flexible Genetic Hybrid Algorithm for nonlinear mixed-integer programming problems. Evol Optim 1(1):41–52
Östermark R (2003) A multipurpose parallel Genetic Hybrid Algorithm for nonlinear nonconvex programming problems. Eur J Oper Res 152:195–214. doi:10.1016/S0377-2217(02)00672-0
Östermark R (2007) A flexible platform for mixed-integer non-linear programming problems. Kybernetes Int J Syst Cybern 36(5/6):652–670
Östermark R (2008a), Geno-mathematical identification of the multi-layer perceptron. Neural Comput Appl (forthcoming)
Östermark R (2008b) Scalability of the Genetic Hybrid Algorithm on a parallel supercomputer. Kybernetes Int J Syst Cybern (forthcoming)
Pukkila T (1980) On transfer function noise model identification. Department of Mathematical Sciences, University of Tampere, Finland (paper presented to the eighth Nordic Conference on Mathematical Statistics held on Åland, Finland 26–29.5)
Pukkila T (1982) On the identification of transfer function noise models with several correlated inputs. Scand J Stat 9:139–146
Rahiala M (1986) Identification and preliminary estimation of linear transfer function models. Scand J Stat 13:239–255
Spliid H (1983) A fast estimation method for the vector autoregressive moving average model with exogenous variables. J Am Stat Assoc 78(384):843–849. doi:10.2307/2288194
Tiao GC, Box GEP (1981) Modelling multiple time series with applications. In: Theory and methods section. J Am Stat Assoc 76(376):802–816
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix 1
VARMA-search for stock prices are given in Table 7.
Appendix 2: Iterative and symplectic estimates of the RICCATI-equation for the covariance matrix of the state vector
(a) Test period 01/86–11/86.
Iterative estimate:
Symplectic estimate:
(b) Test period 01/87–12/88
Iterative estimate:
Symplectic estimate:
Rights and permissions
About this article
Cite this article
Östermark, R. Genetic hybrid tuning of VARMAX and state space algorithms. Soft Comput 14, 91–99 (2010). https://doi.org/10.1007/s00500-008-0393-x
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-008-0393-x