An Application of Markov-Switching Model to Stock Returns Analysis

• Autorzy: Monika Kośko
• Języki publikacji: EN

Abstrakty

EN

The purpose of this paper was to show the application of autoregressive Markov-switching model MS in stock market returns analysis, then research of the properties of this model and its comparison with ARMA(p,q) model - one of the most popular in time series analysis. The empirical results indicate the MSV model, both for weekly and daily data, as the most proper one for description of different parameters structure and relationships between them. All examined states were extremely persistent, what can be a proof for the occurrence of some structure with different parameters (the mean and the variance of residuals), that depends on current change in process of explained variable. No ARCH effect in weekly returns implies better properties of MS models for lower frequency data, in which the ARCH effect is weaker. In the daily data models, the ARCH effect was not eliminated. Therefore, it is said that MS models do not explain the volatility clustering present in the daily stock returns data, what can be preface to further research aiming at construction a Markov-switching ARCH model (SWARCH).(fragment of text)

Słowa kluczowe

EN

Markov models; Switching model; Stock rate of returns

PL

Modele Markowa; Model przełącznikowy; Stopa zwrotu akcji

• Czasopismo: Dynamic Econometric Models
• Rocznik: 2006
• Tom: 7
• Strony: 259--268

Twórcy

autor

Monika Kośko

• The University of Computer Sciences and Economics in Olsztyn, Poland

Bibliografia

• Brzeszczyński, J., Kalm R. (2002), Ekonometryczne modele rynków finansowych, (Econometric models of financial markets), WIG-Press, Warszawa.
• Engle, R. F. (1982), Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of United Kingdom Inflation, Econometrica, vol. 50.
• Garcia, R., Pierre, P. (1996), An Analysis of the Real Interest Rate under Regime Shift, Review of Economics and Statistics, vol. 78.
• Hamilton, J. D. (1989), A New Approach to the Economic Analysis of Nonstationary Time Series and the Busines Cycle, Econometrica, vol. 57.
• Kim, C. J., Nelson, C.R. (1999), State-Space Models with Regime Switching, The MIT Press, London.
• Koskinen, L., Pukkila, T. (1995), An Application of the Vector Autoregressive Model with a Markov Regime to Inflation Rates, źródła internetowe.
• Krolzig, H. M. (1998), Econometric Modelling of Markov Switching Vector Autoregressions using MSVAR for Ox, Institute for Economics and Statistics, Oxford.
• Linne, T. (2002), A Markov Switching Model of Stock Returns: An Application to the Emerging Markets in Central and Eastern Europe, East European Transition and EU Englargement A Quantitative Approach, Berlin.
• Podgórska M. (2002), Łańcuch Markowa w teorii i zastosowaniach, (A Markov chain theory and application), SGH, Warszawa.
• Stawicki, J. (2004), Wykorzystanie łańcuchów Markowa w analizie rynków kapitałowych, (A Markov chain application to capital market analysis), Wydawnictwo UMK, Toruń. Yin, P. (2003), Markov Switching in the Stock Market, Economics, vol. 413, www.missouri.edu/~econprm/ec413f02/pyin.