Acessibilidade / Reportar erro

Previsões Macroeconômicas Baseadas em Modelos TVP-VAR: Evidências Para o Brasil

Vector autoregressive models with time-varying parameters and heteroskedastic effects, proposed in Koop & Korobilis (2013)Koop, G., & Korobilis, D. (2013). Large time-varying parameter VARs. Journal of Econometrics, 177(2), 185-198. doi: 10.1016/j.jeconom.2013.04.007
https://doi.org/10.1016/j.jeconom.2013.0...
and known as TVP-VAR models, are used to predict inflation, the interest rate, and a monthly GDP indicator for several forecast horizons. Dynamic model selection and averaging strategies for the different model specifications are also considered. Forecasts are compared to those from Bayesian VAR models, factor augmented VARs, and other competing models using the model confidence set methodology. The results indicate that TVP-VAR is the only specification always included in the set of best performing models, independently from the variable analyzed, and from the forecast horizon.


Fundação Getúlio Vargas Praia de Botafogo, 190 11º andar, 22253-900 Rio de Janeiro RJ Brazil, Tel.: +55 21 3799-5831 , Fax: +55 21 2553-8821 - Rio de Janeiro - RJ - Brazil
E-mail: rbe@fgv.br