Choice of Variables in Vector Autoregressions∗
نویسندگان
چکیده
Suppose that a dataset with N time series is available. N1 < N of those are the variables of interest. You want to estimate a vector autoregression (VAR) with the variables of interest. Which of the remaining N−N1 variables, if any, should you include in the VAR with the variables of interest? We develop a Bayesian methodology to answer this question. This question arises in most applications of VARs, whether in forecasting or impulse response analysis. We apply the methodology to the euro area data and find that when the variables of interest are the price level, GDP, and the short-term interest rate, the VAR with these variables should also include the unemployment rate, the spread between corporate bonds and government bonds, the purchasing managers index, and the federal funds rate. Of independent interest, we develop Bayesian tests of block-exogeneity – Granger causality – in VARs.
منابع مشابه
Online Appendix to “Granger-Causal-Priority and Choice of Variables in Vector Autoregressions”
In the paper we use the set of models Ω defined in Definition 3 in Section 4. In Section 5.4 we also report findings conditional on the set of models Ω̃. In this online appendix we give the details of the exercise conditional on Ω̃. The motivation for this exercise is the following. The set of models Ω̃ is larger than the set of models Ω. In particular, Ω includes models with one Granger-noncausal...
متن کاملPrior Selection for Vector Autoregressions∗
Vector autoregressions (VARs) are flexible time series models that can capture complex dynamic interrelationships among macroeconomic variables. However, their dense parameterization leads to unstable inference and inaccurate out-ofsample forecasts, particularly for models with many variables. A potential solution to this problem is to use informative priors, in order to shrink the richly param...
متن کاملOn the Identification of Structural Vector Autoregressions
F ollowing seminal work by Sims (1980a, 1980b), the economics profession has become increasingly concerned with studying sources of economic fluctuations. Sims’s use of vector autoregressions (VARs) made it possible to address both the relative importance and the dynamic effect of various shocks on macroeconomic variables. This type of empirical analysis has had at least two important consequen...
متن کاملBayesian Vector Autoregressions
This article provides an introduction to the burgeoning academic literature on Bayesian Vector Autoregressions, benchmark models for applied macroeconomic research. We first explain Bayes’ theorem and the derivation of the closed-form solution for the posterior distribution of the parameters of the model given data. We further consider parameter shrinkage, a distinguishing feature of the prior ...
متن کاملGranger-Causal-Priority and Choice of Variables in Vector Autoregressions∗
A researcher is interested in a set of variables that he wants to model with a vector autoregression and he has a dataset with more variables. Which variables from the dataset to include in the VAR, in addition to the variables of interest? This question arises in many applications of VARs, in prediction and impulse response analysis. We develop a Bayesian methodology to answer this question. W...
متن کامل