方法对比

	局部投影	全球向量自回归模型	时变参数因子增强向量自回归模型
领域	计量经济学	计量经济学	计量经济学
方法族	Regression model	Regression model	Regression model
起源年份	2005	2004	2005
提出者	Oscar Jorda	Pesaran, Schuermann, and Weiner	Bernanke, Boivin, and Eliasz
类型	Multi-horizon regression	International system model	Time-varying system
开创性文献	Jorda, O. (2005). Estimation and inference of impulse responses by local projections. American Economic Review, 95(1), 161-182.	Pesaran, M. H., Schuermann, T., & Weiner, S. M. (2004). Modeling regional interdependencies using a global error-correcting macroeconometric model. Journal of Business and Economic Statistics, 22(2), 129-162.	Bernanke, B. S., Boivin, J., & Eliasz, P. S. (2005). Measuring monetary policy. Journal of Political Economy, 113(1), 161-208.
别名	LP-IR, Multi-horizon regression	GVAR, Multi-country VAR	Dynamic factor model with time-varying parameters
相关	3	3	3
摘要	Local Projections (LP) is a semi-parametric method for estimating impulse responses directly via multi-horizon regressions, bypassing VAR-model specification. Introduced by Jorda (2005), it projects outcomes h periods ahead onto current shocks and lags, producing impulse-response functions without assuming a particular lag structure or VAR order. This flexibility has made it the dominant approach in applied macroeconomics for measuring policy effects and shock transmission.	Global VAR (GVAR) is a large-scale macroeconomic modeling framework linking multiple countries (or regions) via trade and financial channels, allowing shocks in one country to propagate through the global system. Introduced by Pesaran et al. (2004), it solves the curse of dimensionality in international VAR models by estimating country-specific VARs conditional on foreign variables, then solving a system linking all countries. This approach is invaluable for analyzing global spillovers and international policy coordination.	TVP-FAVAR is a hybrid framework combining factor-augmented VARs with time-varying parameter estimation via Kalman filtering. Introduced by Bernanke et al. (2005) and refined by Primiceri (2005), it extracts latent economic factors (e.g., a 'common monetary policy shock') from high-dimensional data while allowing VAR coefficients to evolve stochastically over time. This framework captures both reduced-dimensionality patterns and structural instability, making it ideal for studying evolving policy regimes and shock dynamics.