Working Paper: CEPR ID: DP15411
Authors: Gabriele Fiorentini; Enrique Sentana
Abstract: Likelihood inference in structural vector autoregressions with independent non-Gaussian shocks leads to parametric identification and efficient estimation at the risk of inconsistencies under distributional misspecification. We prove that autoregressive coefficients and (scaled) impact multipliers remain consistent, but the drifts and standard deviations of the shocks are generally inconsistent. Nevertheless, we show consistency when the non-Gaussian log-likelihood is a discrete scale mixture of normals in the symmetric case, or an unrestricted finite mixture more generally. Our simulation exercises compare the efficiency of these estimators to other consistent proposals. Finally, our empirical application looks at dynamic linkages between three popular volatility indices.
Keywords: consistency; finite normal mixtures; pseudo maximum likelihood estimators; structural models; volatility indices
JEL Codes: C32; C46; C51; C58
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| estimators are efficient compared to other consistent proposals (C51) | validation of proposed estimators (C52) |
| maximum likelihood estimation in SVAR models with independent non-Gaussian shocks (C51) | consistent estimators for autoregressive coefficients and scaled impact multipliers (C51) |
| maximum likelihood estimation in SVAR models with independent non-Gaussian shocks (C51) | inconsistencies in estimating the drifts and standard deviations of the shocks (C51) |
| non-Gaussian log-likelihood as a discrete scale mixture of normals (C46) | all parameters remain consistently estimated (C51) |
| structural shocks (E32) | impact on volatility measures (C58) |