faty
Get Complete Project Material File(s) Now! »
VECX* models
VECX* models are also classified as cointegrating VARX or cointegrating VARX* models (Affandi, 2007; Garratt et al., 2006). Pesaran et al. (2000), Pesaran and Shin (2002) and Garratt et al. (2003) introduced and further developed these models. A detailed explanation of the methodology is provided in Garratt et al. (2006). Pesaran et al. (2000) explain that the models are particularly suitable for small open economies due to the handling of foreign variables as weakly exogenous. Pesaran and Smith (2006) further illustrate that this type of model can be derived as the solution to an open macro economy New Keynesian dynamic stochastic general equilibrium (DSGE) model, thereby underpinning the long-run relations considered in the VECX* model. Therefore, the first advantage of using a VECX* approach for South Africa is that the model accounts for long-run theoretical relations and short-run properties, which are both important in the analysis of the impact of monetary policy shocks on the system. Second, the inclusion of weakly exogenous foreign variables, which is relevant for a small open economy such as South Africa, is possible in a VECX* model. Another advantage of developing a VECX* for South Africa is that it can be incorporated directly into a GVAR model, where all the foreign variables are determined endogenously. Pesaran, Schuermann and Weiner (2004) proposed the GVAR framework.
VECX* MODEL RESULTS
The variables included in the VECX* model are based on the results of the preliminary data analysis. As indicated in Table 2.1, the domestic endogenous variables included in the VECX* model are real output ( y), the quarterly inflation rate (π), the repo rate (r ), the long-term interest rate (lr) and the real effective exchange rate (calculated as e p ep , from the nominal effective exchange rate and prices). The weakly exogenous variables are foreign real output ( * y ), foreign prices ( * p ), the foreign short-term interest rate ( r ) and the oil price ( oil p ).The VECX* model for South Africa was developed in Microfit 5.0 (Pesaran & Pesaran, 2009a; 2009b). The Akaike information criterion (AIC) indicates that the optimal model has two lags of the endogenous variables and one lag of the exogenous variables. Table 2.6 contains the cointegration test results for the model with an unrestricted intercept, a restricted trend, a restricted dummy variable to account for shifts in long-run relations due to the structural change in South Africa in the early 1990s (d92 ), and an unrestricted differenced dummy variable to allow for shifts in short-run dynamics due to the structural change ( ( ) d92D ). A deterministic trend is included since most of the variables are trended; however, the trend is restricted to lie within the cointegrating space to avoid the possibility of quadratic trends in the solution of the model in levels. The marginal models for the weakly exogenous foreign variables each include one lag for the differenced endogenous variables, one lag for the differenced exogenous variables and an intercept. In Table 2.6, the maximum eigenvalue statistic suggests a rank of three at a 10 per cent level of significance, while the trace statistic indicates a rank of three at a five per cent level of significance. Thus, there are three cointegrating relationships in the model, which are in line with the three significant long-run economic relations identified in Section 2.5.
CHAPTER 1 INTRODUCTION
1.1. BACKGROUND
1.2. OUTLINE OF STUDY
CHAPTER 2 MONETARY POLICY AND INFLATION IN SOUTH AFRICA: A VECM AUGMENTED WITH FOREIGN VARIABLES
2.1. INTRODUCTION
2.2. LITERATURE REVIEW
2.3. VECX* METHODOLOGY
2.4. DATA
2.5. DATA ANALYSIS
2.6. VECX* MODEL RESULTS
2.7. CONCLUSION
CHAPTER 3 THE IMPACT OF ECONOMIC SHOCKS IN THE REST OF THE WORLD ON SOUTH AFRICA: EVIDENCE FROM A GLOBAL VAR MODEL
3.1. INTRODUCTION
3.2. LITERATURE REVIEW
3.3. GVAR METHODOLOGY
3.4. GVAR SPECIFICATION AND EMPIRICAL ESTIMATION
3.5. RESULTS OF SHOCKS TO THE GVAR
3.6. CONCLUSION
CHAPTER 4 FORECASTING KEY SOUTH AFRICAN VARIABLES WITH A GLOBAL VAR MODEL
4.1. INTRODUCTION
4.2. LITERATURE REVIEW
4.3. METHODOLOGY
4.4. MODEL SPECIFICATIONS
4.5. FORECAST EVALUATION
4.6. CONCLUSION
CHAPTER 5 GENERAL CONCLUSION AND AREAS OF FUTURE RESEARCH
5.1. INTRODUCTION
5.2. CONTRIBUTIONS OF THIS STUDY
5.3. SUMMARY OF KEY FINDINGS
5.4. AREAS OF FUTURE RESEARCH
REFERENCES
