Comparer des méthodes

	Modèle VAR Structurel à Paramètres Variant dans le Temps (TVP-SVAR)	Modèle VAR bayésien (BVAR)
Domaine	Économétrie	Économétrie
Famille	Regression model	Regression model
Année d'origine	2005	1984
Auteur d'origine	Giorgio E. Primiceri	Doan, Litterman & Sims
Type	Bayesian state-space SVAR	Multivariate time-series model
Source fondatrice	Primiceri, G. E. (2005). Time varying structural vector autoregressions and monetary policy. Review of Economic Studies, 72(3), 821–852.	Doan, T., Litterman, R., & Sims, C. (1984). Forecasting and conditional projection using realistic prior distributions. Econometric Reviews, 3(1), 1–100.
Alias	TVP-SVAR, time-varying SVAR, drifting-parameter SVAR, TVP structural VAR	BVAR, Bayesian VAR, Bayesian vector autoregressive model, BVAR model
Apparentées	2	5
Résumé	The Time-Varying Parameter Structural VAR (TVP-SVAR) model extends classical structural VARs by allowing both the reduced-form coefficients and the structural impact matrix to evolve continuously over time. Estimated via Bayesian MCMC, it captures shifting transmission mechanisms and heteroscedastic volatility — making it the workhorse for empirical macroeconomics when policy regimes and economic relationships change.	The Bayesian Vector Autoregression (BVAR) model extends the classical VAR framework by incorporating prior beliefs about the model coefficients. Priors — most commonly the Minnesota prior — shrink VAR coefficients toward economically sensible values, dramatically reducing overfitting and improving out-of-sample forecast accuracy even when the number of variables is large.