Small‐sample confidence intervals for multivariate impulse response functions at long horizons

Abstract

SUMMARYExisting methods for constructing confidence bands for multivariate impulse response functions may have poor coverage at long lead times when variables are highly persistent. The goal of this paper is to propose a simple method that is not pointwise and that is robust to the presence of highly persistent processes. We use approximations based on local-to-unity asymptotic theory, and allow the horizon to be a fixed fraction of the sample size. We show that our method has better coverage properties at long horizons than existing methods, and may provide different economic conclusions in empirical applications. We also propose a modification of this method which has good coverage properties at both short and long horizons.

Gimme a Break! Identification and Estimation of the Macroeconomic Effects of Monetary Policy Shocks in the U.S.

“…:::!A k s k , has exactly n ! r unit roots equal to s=1, with 0 " r < n (recall that n:=7) 2 Then consider the 1 Alternatively, it is generally possible to treat the system as a highly persistent one and avoid pre-testing by constructing IRFs using the local-to-unity asymptotic theory as in, e.g., Pesavento and Rossi (2006). 2 We recall that NDCONS stands for non-durable personal consumption, DCONS for durable peri cointegrated Vector Equilibrium Correction (VEqC) counterpart of the system:…”

One-Dimensional Inference in Autoregressive Models With the Potential Presence of a Unit Root

“…That is, this asymptotic distribution does not depend on any nuisance parameters, and the usual bootstrap will work (Inoue and Kilian (2002)). However, if k T = [δT ], the horizon proportionally increases with the sample size as in Gospodinov (2004) and Pesavento and Rossi (2006), then the ratio of…”

“…The comprehensive simulation study of Pesavento and Rossi (2007) compared coverage (size) of classical methods for performing inference on impulse responses such as the deltamethod, pretest and bootstrap, as well as methods based on a linearly-growing horizon such as Wright (2000), Gospodinov (2004) and Pesavento and Rossi (2006). Many of the observations of Pesavento and Rossi (2007) are explained and theoretically justified in previous sections.…”

“…The assumption of a growing horizon may be a reasonable one for the estimation of an impulse responses on long horizons. Methods suggested by Gospodinov (2004) and Pesavento and Rossi (2006) employ this asymptotic setting. The confidence sets based on the classical methods (delta-method, bootstrap) are significantly different from those produced by methods that are consistent under linearly-growing horizon asymptotics.…”

Forecasting Bank Stock Market Prices With a Hybrid Method: The Case of Alpha Bank / Vertybinių Popierių Kainų Prognozavimas Hibridiniu Metodu: Alpha Bank Pavyzdys

“…of the predictions. The technique of bootstrapping applied on the residuals has been extensively used in the past (Shao, Tu 1955;Efron, Tibshirani 1986;Hall 1986Hall , 1988Beran 1988;Franklin, Wasserman 1992;Simar, Wilso n 1998;Glaz, Sison 1999;Bjørnstad, Falck 2001;Tribouley 2004;Chou 2006;Pesavento, Rossi 2006;Kapetanios 2008;Xiong, Li 2008;Charitos et al 2009;Jun Li et al 2009;Annaert et al 2009;Kascha, Mertens 2009;Barnes et al 2009). 4 th step: Based on the process mentioned above two new time series of the upper and lower limits of the B.C.I.…”