Quantitative Macroeconomic Modeling with ... - EViews

1 Quantitative Macroeconomic Modeling withStructural Vector Autoregressions An EViewsImplementationS. Ouliaris1, Pagan2and J. Restrepo3 September 19, 2016 S. Ouliaris, Pagan and J. Restrepo. All rights book began as a series of lectures given by the second author at the In-ternational Monetary Fund as part of the Internal Economics Training programconducted by the Institute for Capacity Development. They continued for theyears from 2011-2015 and were gradually adapted to describe the methods avail-able for the analysis of Quantitative Macroeconomic systems with the StructuralVector Autoregression approach. A choice had to be made about the computerpackage that would be used to perform the Quantitative work and EViews waseventually selected because of its popularity amongst IMF staff and centralbankers more generally. Although the methodology developed in this book ex-tends to other packages such as Stata, it was decided to illustrate the methodswith the book developed out of a set of lectures we would wish to thankthe many IMF staff and country officials who participated in the courses andwhose reactions were important in allowing us to decide on what should beemphasised and what might be treated more lightly.

2 The courses were excep-tionally well organized by Luz Minaya and Maria (Didi) Jones. Versions of thecourse were also given at the Bank of England and the Reserve Bank of Aus-tralia. We also need to thank Gareth Thomas and Glenn Sueyoshi at IHS GlobalInc for their help in understanding some of the functions of the EViews packageand for providing some new options in EViews that we found important forimplementing the methods of this , on a personal level Adrian would like to dedicate the book to Janetwho had to endure the many hours of its construction and An Overview of Macro-econometric System Modeling132 Vector Autoregressions: Basic Basic Structure .. Likelihood Estimation of Basic VARs .. Small Macro Model Example .. Specification of VARs .. models .. of Thumb .. Criteria .. of Variables .. Knowledge .. models.

3 VAR s .. Some Lag Coefficients to Zero .. Exogeneity- the VARX Model .. VARs .. and Dummy Variables .. Factors .. Conclusion .. 393 Using and Generalizing a Introduction .. Testing Granger Causality .. Forecasting using a VAR .. Evaluation .. Forecasts .. Using EViews .. Bayesian VARs .. Minnesota prior.. the Minnesota prior in EViews .. prior.. the Normal-Wishart prior in EViews priors using dummy observations or pseudo data of Coefficients Dummy Prior .. Initial Observations Dummy Prior .. with Bayesian VARs .. Computing Impulse Responses .. Standard Errors for Impulse Responses .. Issues when Using the VAR as a Summative Model .. Variables .. Variables .. VARs .. Switching process .. Varying VARs.

4 Variables for Recurrent States .. 744 Structural Vector Autoregressions with I(0) Introduction .. Mathematical Approaches to Finding Uncorrelated Shocks .. Generalized Impulse Responses .. Structural VAR s and Uncorrelated Shocks: Representation andEstimation .. Likelihood Estimation .. Variable (IV) Estimation .. Impulse Responses for an SVAR: Their Construction and Use .. and Variable Decompositions .. Restrictions on an SVAR .. Systems .. Recursive SVAR with the US Macro Data .. the Recursive Small Macro Modelwith EViews .. Response Anomalies (Puzzles) .. Restrictions on the Impact of Shocks .. Zero Contemporaneous Restriction .. Two Periods Ahead Zero Restriction .. Restrictions on Parameters - The Blanchard-Perotti Fiscal Policy Model.

5 Stocks and Flows Plus Identities into anSVAR - A US Fiscal-Debt Model .. Exogenous Variables in an SVAR - the SVARXM odel .. on Parameters and Partial Exogeneity: Ex-ternal Instruments .. Augmented SVARs .. SVARs (SGVARs) .. Models and the Origins of SVARs .. Standard Errors for Structural Impulse Responses.. Other Estimation Methods for SVARs .. Higher Order Moment Information .. 1355 SVARs with I(0) Variables and Sign Introduction .. The Simple Structural Models Again and Their Sign Restrictions How Do we Use Sign Restriction Information? .. SRR Method .. the Orthogonal Matrices .. SRC method .. SRC and SRR Methods Applied to a Market Model . SRR Method Applied to the Market Model SRC Method Applied to the Market Model the SRR and SRC Methodologies .. SRC and SRR With a Simulated Market SRC and SRR With a Small Macro Model andTransitory Shocks.

6 What Can and Can t Sign Restrictions Do for You? .. Restrictions Will Not Give You a Single Model - TheMultiple Models Problem .. Restrictions and the Size of Shocks? .. Do the True Impulse Responses Lie in the Rangeof Generated Models? .. Do We Do About Multiple Shocks? .. Can Sign Restrictions do for you? .. Sign Restrictions in Systems with Block Exogeneity .. Standard Errors for Sign Restricted Impulses .. SRR Method .. SRC method .. Summary .. 1606 Modeling SVARs with Permanent and Transitory Introduction .. Variables and Shocks .. Why Can t We Use Transitory Components ofI(1) Variables inSVARs? .. SVARs with Non-CointegratedI(1) andI(0) Variables .. Two Variable System inI(1) Variables .. EViews Application of the Two I(1) VariableModel .. Alternative EViews Application of the TwoI(1) Variable Model.

7 Two Variable System with a Permanent and TransitoryShock - the Blanchard and Quah Application in EViews . Solution for the Two Variable Case .. Four Variable Model with Permanent Shocks - Peers-man (2005) .. the Small Macro Model with a Permanent Sup-ply Shock .. Problems with Measuring Uncertainty in Impulse Responses .. Sign Restrictions when there are Permanent and Transitory Shocks1997 SVARs with Cointegrated and I(0) Introduction .. The VECM and Structural VECM Models .. SVAR Forms of the SVECM .. and Transitory Shocks Only .. , Transitory and Mixed Shocks .. Example: Gali s (1999) Technology Shocks and Fluctuations of System and Restrictions Used .. of the System .. Impulse Responses to a Single Shock .. Example: Gali s 1992 IS/LM Model .. of System and Restrictions Used .. of the System.

8 215 List of Opening A Workfile in EViews 9 .. Chomoreno Data Set in EViews 9 .. Specification of the Small Macro Model (VAR) in EViews 9 .. Results from Fitting a VAR(2) to the Small Macro Model Data Set EViews Program to Replicate the Small Macro Model Results .. Choice of VAR Lag Length for the Small Macro Model .. Specification of a VARX Model for the Brazilian Macro Data Brazilian VAR Results, 1992:2 2008:4 .. Moving to the System Estimator in EViews .. System Representation of the Brazilian VAR with ExogeneityRestrictions .. Ordinary Least Squares Estimates of the Restricted VARX Model Screen Shot of the Options for Testing Breaks in the Coefficientsof the VAR .. Standardized Variables from the Bernankeet al.(2005) Data Extracting Three Principal Components from the Bernankeet al.

9 (2005) Data Set .. Forecasting a Reduced Form VAR using EViews : Direct Approach EViews Output: Forecast Tab .. Creating a VAR model using EViews .. VAR Model Object: Chomoreno .. Generating a VAR Forecast Using the VAR Model Object .. Federal Funds Rate (FF) Under the Alternative Scenario (Sce-nario 1) .. Conditional Forecasting Using the EViews Model Simulator: Edit-ing the Alternative Scenario .. Conditional Forecasting Using the EViews Model Simulator: Over-riding a Variable .. Simulating the Chomoreno model Under An Alternative Scenario Conditional Forecasts for GAP and INFL using the ChomorenoVAR .. Estimating a Bayesian VAR using the EViews VAR Object .. Bayesian VAR Estimation in EViews : Prior type Dialog Box . 527 LIST OF Bayesian VAR Estimation in EViews : Prior specification Dia-log Box.

10 VAR versus BVAR Estimates (Minnesota Prior): ChomoronoModel, 1981Q3-1997Q4.. Selecting the Normal-Wishart Prior in EViews .. Specifying the Hyper-Parameters for the Normal-Wishart Prior . Bayesian VAR Estimates using a Normal-Wishart Prior .. Generating Impulse Responses in EViews .. Types of Impulse Shocks in EViews .. Impulse Responses of the Output Gap, Inflation and the IntereestRate to a Unit Change in the VAR Output Gap Equation Impulse Responses for the Recursive Small Macro Model .. WritingAe(t) =Bu(t) in EViews .. MLE Estimation of A and B matrices for the Small StructuralModel Using EViews .. Example of a Matrix Object in EViews .. Example an A Matrix for the Recursive Small Macro Model .. Creating a System Object CalledchomorsysUsing EViews .. Specification of the Small Macro Model in an EViews SYSTEMO bject.