目录
Contents
Foreword vii
Chapter 1: From VAR models to Structural
VAR models 1
1.1. Origins of VAR modelling 1
1.2. Basic concepts of VAR analysis 2
1.3. Efficient estimation: the BVAR approach 6
1.4. Uses of VAR models 10
1.4.1. Dynamic simulation 10
1.4.2. Unconditional and conditional forecasting 11
1.4.3. Granger causality 13
1.5. Different classes of Structural VAR models 15
1.6. The likelihood function for SVAR models 19
1.7. Structural VAR models vs. dynamic
simultaneous equations models 22
1.8. Some examples of Structural VARs in the
applied literature 23
1.8.1. Triangular representation deriving from the
Choleski decomposition of Σ 24
1.8.2. Blanchard and Quah (1989) long run constraints 24
1.8.3. A traditional interpretation of macroeconomic
fluctuations: Blanchard (1989) 26
Chapter 2: Identification analysis and F.I.M.L.
estimation for the K-Model 29
2.1. Identification analysis 29
2.2. F.I.M.L. estimation 36
Chapter 3: Identification analysis and F.I.M.L.
estimation for the C-Model 40
3.1. Identification analysis 40
3.2. F.I.M.L. estimation 45
Chapter 4: Identification analysis and F.I.M.L.
estimation for the AB-Model 48
4.1. Identification analysis 48
4.2. F.I.M.L. estimation 57
Chapter 5: Impulse response analysis and forecast error
variance decomposition in SVAR modelling 60
5.1. Impulse response analysis 60
5.2. Variance decomposition (by Antonio Lanzarotti) 67
5.3. Finite sample and asymptotic distributions for
dynamic simulations 73
Chapter 6: Long run a priori information.
Deterministic components. Cointegration 78
6.1. Long run a priori information 78
6.2. Deterministic components 82
6.3. Cointegration 85
6.3.1. Representation and identification issues 88
6.3.2. Estimation issues 91
6.3.3. Interpretation of the cointegrating coefficients 98
6.3.4. Asymptotic distributions of the parameter estimates:
Structural VAR analysis with cointegrated series 100
6.3.5. Finite sample properties 103
Chapter 7: Model selection in Structural VAR analysis 107
7.1. General aspects of the model selection problem 107
7.2. The dominance ordering criterion 108
7.3. The likelihood dominance criterion (LDC) 111
Chapter 8: The problem of non fundamental
representations 114
8.1. Non fundamental representations in time series models 114
8.2. Economic significance of non fundamental
representations and examples 118
8.3. Non fundamental representations and applied SVAR
analysis 120
8.4. An example 125
Chapter 9: Two applications of Structural VAR
analysis 131
9.1. A traditional interpretation of Italian macroeconomic
fluctuations 131
9.1.1. The reduced form VAR model 132
9.1.2. Cointegration properties 133
9.1.3. Structural identification of instantaneous
relationships 134
9.1.4. Dynamic simulation 135
9.2. The transmission mechanism among Italian
interest rates 136
9.2.1. The choice of the variables 136
9.2.2. The reduced form VAR model 137
9.2.3. Cointegration properties 139
9.2.4. Structural identification of instantaneous
relationships 143
9.2.5. Dynamic simulation 145
9.2.6. The Lippi-Reichlin criticism 149
Annex 1: The notions of reduced form and structure in
Structural VAR modelling 151
Annex 2: Some considerations on the semantics, choice
and management of the K, C, and AB-models 154
Appendix A 159
Appendix B 162
Appendix C (by Antonio Lanzarotti and
Mario Seghelini) 165
References 174