ARTICLE
ABSTRACT
The paper presents the three different stochastic unit root models (STUR), proposed by Granger and Swanson (1997), by Leybourne, McCabe and Mills (1996) and finally by Francq, Makarova, Zakoian (2008). The main purpose is to develop an MCMC algorithm for Bayesian estimation. We are also interested how the different specifications of the stochastic unit root affect the explanatory power of a set of competing models. We apply that these computational methods to daily exchange rates of foreign currencies in zlotys, namely Swiss franc, Pound sterling and Euro. The model selection and posterior estimates provide strong evidence in favor of the STUR models.
REFERENCES
[1] Bruzda J. (2003), Procesy dwuliniowe i procesy GARCH w modelowaniu finansowych szeregów czasowych, Przegląd Statystyczny, 2, 73-95.
[2] Doman M., Doman R. (2004), Ekonometryczne modelowanie dynamiki polskiego rynku finansowego, Wyd. AE w Poznaniu, Poznań.
[3] Francq Ch., Makarova S., Zako¨ian J.M. (2008), A class of stochastic unit-root bilinear processes: Mixing properties and unit-root test, Journal of Econometrics, 142, 312-326.
[4] Granger W.J.C., Andersen A. P. (1978), An Introduction to Bilinear Time Series Models, Vandenhoeck and Ruprecht, Göttingen.
[5] Granger C.W.J., Swanson N.R. (1997), An Introduction to stochastic unit–root process, Journal of Econometrics, 80, 35-62.
[6] Granger W.J.C., Teräsvirta T. (1993), Modeling Nonlinear Economic Relationships, Oxford University Press, Oxford.
[7] Górka J. (2007), Modele autoregresyjne z losowymi parametrami, Procesy STUR. Modelowanie i zastosowanie do finansowych szeregów czasowych, red. Osińska, M., Wyd. TNOiK, Toruń, 13-34.
[8] Jones C.R., Marriott J.M. (1999), A Bayesian analysis of stochastic unit root models, Bayesian Statistics, 6, 785-794.
[9] Koop G. (2003), Bayesian Econometrics, John Wiley & Sons.
[10] Leybourne S.J., McCabe B.P.M., Mills T.C. (1996), Randomized unit root processes for modelling and forecasting financial time series: theory and applications, Journal of Forecasting, 15, 253-270.
[11] Leybourne S.J., McCabe B.P.M., Tremayne A.R. (1996), Can economic time series be differenced to stationarity? Journal of Business and Economic Statistics, 14, 435-446.
[12] Newton M.A., Raftery A.E. (1994), Approximate Bayesian inference by the weighted likelihood bootstrap (with discussion), Journal of the Royal Statistical Society, B, 56, 3-48.
[13] Nicholls D.F., Quinn B.G. (1982), Random Coefficient Autoregressive Models: An Introduction, Springer-Verlag, New York.
[14] Osiewalski J., Pajor A., Pipien M. (2004), Bayesowskie modelowanie i prognozowanie indeksu WIG z wykorzystaniem procesów GARCH i SV, XX Seminarium Ekonometryczne im. Profesora Zbigniewa Pawlowskiego, red. Zelias, A., Wyd. AE w Krakowie, Kraków, 17-39.
[15] Osińska M. (red) (2007), Procesy STUR. Modelowanie i zastosowanie do finansowych szeregów czasowych, Wyd. TNOiK, Toruń.
[16] Osińska M. (2004), Stochastic unit roots process – properties and application, 30-rd International Conference MACROMODELS’04, red. Welfe A., Welfe W., Wyd. UŁ w Łodzi, Łódź, 169-179.
[17] Osińska M., Górka J. (2005), Identyfikacja nieliniowości w ekonomicznych szeregach czasowych. Analiza symulacyjna, Dynamiczne modele ekonometryczne, Toruń, 35-44.
[18] Pajor A. (2003), Procesy zmienności stochastycznej SV w bayesowskiej analizie finansowych szeregów czasowych, Wyd. AE w Krakowie, Kraków.
[19] Sollis R., Leybourne S.J., Newbold P. (2000), Stochastic unit roots modelling of stock price indices, Applied Financial Economics, 10, 311-315.
[20] Tsay R.S. (2005), Analysis of Financial Time Series, Wiley, John and Sons, Inc., Hoboken, New Jersey.
[21] Yoon G. (2006), A note on some properties of STUR processes, Oxford Bulletin of Economics and Statistics, 68, 2, 253-260.
