Warianty tytułu
Bayesian Analysis of Autoregression Models and a Mobile Average
Języki publikacji
Abstrakty
Modele typu ARIMA, wprowadzone przez Boxa i Jenkinsa (1976), są przydatne w analizie szeregów czasowych do opisu wielu zmiennych ekonomicznych. W pracy przedstawiono bayesowską procedurę identyfikacji i estymacji modeli typu ARMA. Opisane metody wykorzystano następnie do analizy danych empirycznych. (abstrakt oryginalny)
Model choice and assessment are a crucial part of statistical data analysis. Due to recent computational advances, sophisticated techniques for Bayesian model comparison are becoming increasingly popular. The paper discuses Bayesian identification and estimation of ARMA (p, q) models. In practice more processes have a low order (p + q ≤ 2) , therefore we take into consideration only the choice from among 5 alternative models: ARMA (1,0 ), ARMA (0,1), ARMA (2,0), ARMA (0,2), ARMA (1,1). We apply the following Bayesian criteria of model choice: choose the model for which a posteriori probability of choosing the model is largest. Depending on the order of the model, a different prior distributions of model parameters were specified. The methods presented were implemented for time series selected from the well known text by Box and Jenkins (1976). We also illustrate the effects on Bayesian model selection and estimation of differing sets of prior parameters and different sample sizes. (original abstract)
Twórcy
autor
- Akademia Ekonomiczna im. Karola Adamieckiego w Katowicach
Bibliografia
- [1] BERGER J.O., Statistical Decision Theory and Bayesian Analysis, Springer-Verlag, New York, Inc., 1985.
- [2] BOX G.E.P., JENKINS G.M., Analiza szeregów czasowych, Warszawa, PWN, 1983.
- [3] DORFMAN J.H., Bayesian Economic Through Numerical Methods, Springer-Verlag, New York, Inc. 1997.
- [4] GRANGER C.W.J., NEWBOLD P., Forecasting Economic Time Series, Academic Press, Inc., Harcout Brace Jovanowich, Publishers, 1986.
- [5] MONAHAN J.F., Fully Bayesian Analysis of ARMA Series models, Journal of Econometrics 21, 307-331, 1983.
- [6] SIMS C.A., UHLIG H., Understanding Unit Roots, Econometrica 59, 1591-1600, 1991.
Typ dokumentu
Bibliografia
Identyfikatory
Identyfikator YADDA
bwmeta1.element.ekon-element-000171602237