|
on Econometric Time Series |
By: | Helmut Lütkepohl; Till Strohsal |
Abstract: | This paper analyzes possibly time-varying shock transmission in structural vector autoregressive (VAR) models when the reduced-form VAR coefficients are time-invariant and the shocks are identified through non-Gaussianity. To check for possible time-variation in the impulse responses, we propose Wald tests for two situations: (1) homoskedastic and (2) heteroskedastic structural shocks. For the latter case, the challenge is to ensure that the test does not indicate time-varying impulse responses if the changes are due only to changes in the variances of the shocks. To illustrate the usefulness of the tests, they are applied to an empirical model of the crude oil market. They support time-varying shock transmission reflected in impulse response functions that change over time. |
Keywords: | Structural vector autoregression, independent component analysis, non-Gaussian shocks, structural break tests, heteroskedasticity |
JEL: | C32 |
Date: | 2025 |
URL: | https://d.repec.org/n?u=RePEc:diw:diwwpp:dp2110 |
By: | Lutz Kilian |
Abstract: | Structural impulse response functions may be estimated based on priors about the parameters of the structural VAR presentation. Even when such priors appear seemingly reasonable, they may imply an unintentionally informative prior for the structural impulse responses. Rather than pretending that the posterior of the impulse responses does not depend on this prior, the proposal in this paper is to verify that the prior distribution of the vector of impulse responses of interest is not unintentionally informative. Moreover, if the impulse response prior is intentionally informative, this point must be conveyed, so the reader can properly evaluate the reported conclusions. This paper discusses easy-to-use diagnostic tools that help practitioners address these concerns. |
Keywords: | VAR; prior; posterior; impulse response; inference |
JEL: | C11 C32 C52 Q43 |
Date: | 2025–02–25 |
URL: | https://d.repec.org/n?u=RePEc:fip:feddwp:99620 |
By: | Yuying Sun (School of Economics and Management, University of Chinese Academy of Sciences and Academy of Mathematics and Systems Science, Chinese Academy of Sciences, China); Shaoxin Hong (Center for Economic Research, Shandong University, Jinan, Shandong 250100, China); Zongwu Cai (Department of Economics, The University of Kansas, Lawrence, KS 66045, USA) |
Abstract: | This paper proposes a novel state-varying model averaging prediction for varyingcoefficient models that accounts for parameter uncertainty and model misspecification. We develop a leave-h-out state-dependent forward-validation criterion to select state-varying combination weights. It is shown that the proposed averaging prediction is asymptotically optimal in the sense of achieving the lowest possible out-of-sample prediction risk in a class of model averaging estimators. This complements existing model averaging methods that primarily focus on minimizing the in-sample squared error loss. Besides, when the set of candidate models includes correctly specified models, the proposed approach asymptotically assigns full weight to these models. Furthermore, the proposed approach is flexible and encompasses special cases including ultra-high dimensional models as well as state-varying factor-augmented regression models. Simulation studies and empirical applications highlight the merits of the proposed averaging prediction relative to other existing model averaging and model selection methods. |
Keywords: | Asymptotic optimality; Varying-coefficient models; Forward-validation; Model averaging; Weight convergence |
JEL: | C2 C13 |
Date: | 2025–03 |
URL: | https://d.repec.org/n?u=RePEc:kan:wpaper:202507 |