JEŘÁBEK, Tomáš and Radka ŠPERKOVÁ. A Predictive Likelihood Approach to Bayesian Averaging. Acta Univ Agric Silvic Mendel Brun. Praha: Mendel University Press, 2015, vol. 63, No 4, p. 1269-1276. ISSN 1211-8516. Available from: https://dx.doi.org/10.11118/actaun201563041269.
Other formats:   BibTeX LaTeX RIS
Basic information
Original name A Predictive Likelihood Approach to Bayesian Averaging
Authors JEŘÁBEK, Tomáš (203 Czech Republic, guarantor, belonging to the institution) and Radka ŠPERKOVÁ (203 Czech Republic).
Edition Acta Univ Agric Silvic Mendel Brun, Praha, Mendel University Press, 2015, 1211-8516.
Other information
Original language English
Type of outcome Article in a journal
Field of Study 50200 5.2 Economics and Business
Country of publisher Czech Republic
Confidentiality degree is not subject to a state or trade secret
RIV identification code RIV/04274644:_____/15:#0000041
Organization unit University of Finance and Administration
Doi http://dx.doi.org/10.11118/actaun201563041269
Keywords (in Czech) predictive likelihood; density forecasts; Bayesian averaging; Bayesian VAR model
Keywords in English predictive likelihood; density forecasts; Bayesian averaging; Bayesian VAR model
Tags AR 2016-2017, RIV_ne, SCOPUS, xJ3
Tags Reviewed
Changed by Changed by: Mgr. Tomáš Jeřábek, Ph.D., MBA, učo 29123. Changed: 8/6/2018 10:01.
Abstract
Multivariate time series forecasting is applied in a wide range of economic activities related to regional competitiveness and is the basis of risk management analysis. In this paper we combine multivariate density forecasts. The performance of models is identified using historical dates including domestic economy and foreign economy, which is represented by countries of the Eurozone. Because forecast accuracy of observed models are different, the weighting scheme based on the predictive likelihood, the trace of past MSE matrix, model ranks are used to combine the models. The equal-weight scheme is used as a simple combination scheme. The results show that optimally combined densities are comparable to the best individual models.
Abstract (in Czech)
Multivariate time series forecasting is applied in a wide range of economic activities related to regional competitiveness and is the basis of risk management analysis. In this paper we combine multivariate density forecasts. The performance of models is identified using historical dates including domestic economy and foreign economy, which is represented by countries of the Eurozone. Because forecast accuracy of observed models are different, the weighting scheme based on the predictive likelihood, the trace of past MSE matrix, model ranks are used to combine the models. The equal-weight scheme is used as a simple combination scheme. The results show that optimally combined densities are comparable to the best individual models.
PrintDisplayed: 25/4/2024 05:33