Identifying UK aggregate demand and aggregate supply relations within the long-run structural VAR framework
This thesis is inspired by the ESRC-Cambridge project "Structural Modelling of the UK Economy within a VAR Framework using Quarterly and Monthly Data" and, in particular, the studies by Garratt et al (1998, 2001). The primary aim is to apply the Long-Run Structural Cointegrating VAR approach, developed within the ESRC-Cambridge project, in order to empirically investigate UK Aggregate Demand and Supply. The empirical analysis is intended to complement the recently developed macro-econometric model of the UK in Garratt et al (1998, 2001) by (i) addressing the issue of structural change and (ii) providing an explicit model of the supply-side of the economy. The recently developed techniques in Johansen and Nielsen (1994), Hansen (2000) and Johansen, Mosconi and Nielsen (2000) are utilised in order to control for and assess the possible long-run effects of different exchange rate regimes. In the light of the well documented finite-sample bias, statistical inference relies in large part on simulation methods along the lines of, inter alia, van Giersbergen (1996), Li and Maddala 1997 Harris , and Judge (1998), Mantalos and Shukur (1998), Gredenhoff and Jacobson (1998), Fachin (2000), Jacobson et al (2001) and Greenslade et al (2002). A practical problem concerning the use of these methods for inference on the cointegrating parameters is identified and a solution is proposed. The Generalised Impulse Responses developed in Koop et al (1996) and Pesaran and Shin (1998) and the Persistence Profiles proposed by Lee et al (1992) and Lee and Pesaran (1993a) are used in order to illustrate the dynamic properties of the estimated systems and provide an informal comparison with the Garratt et al (1998, 2001) models.