The South African business cycle and the application of dynamic stochastic general equilibrium models

Date
2014-12
Authors
Kotze, Kevin Lawrence
Journal Title
Journal ISSN
Volume Title
Publisher
Stellenbosch : Stellenbosch University
Abstract
ENGLISH ABSTRACT: This dissertation considers the use of Dynamic Stochastic General Equilibrium (DSGE) models for the analysis of South African macroeconomic business cycle phenomena. It includes four separate, but interrelated parts, which follow a logical sequence. The rst part motivates the use of these models before establishing the theoretical foundations for these models. The theoretical foundations are accompanied by detailed derivations that are used to construct a model for a small open economy. The second part considers the properties of South African macroeconomic data that may be used to estimate the parameters in these models. It includes a discussion of the variables that may be included in such a model, as well as various methods that may be used to extract the business cycle. Thereafter, the sample size for the dataset is established, after investigating for possible structural breaks in the rst two moments of the data, using various univariate and multivariate techniques. The nal chapter of this part contains an investigation into the measures of core in ation, whereby a comparison of trimmed means, dynamic factor models and various wavelet decompositions are applied to data for South Africa. The third part considers the application of the dataset that was identi ed in part two, in a DSGE model that incorporates features that are typical of small open economies. It includes a discussion that relates to the role of the exchange rate in these models, which is found to contain key information. In addition, this part also includes a optimal policy investigation, which considers the reaction function of central bank. The nal part of this thesis considers more recent advances that have been applied to DSGE models for the South African economy. It includes an example of a nonlinear model that is estimated with the aid of a particle lter, which is then used for forecasting purposes. The forecasting results of both linear and nonlinear versions of the model are then compared with the results from various Vector Autoregression (VAR) and Bayesian VAR models.
AFRIKAANSE OPSOMMING: Hierdie proefskrif oorweeg die gebruik van Dinamiese Stogastiese Algemene Ewewig (Engels: Dynamic Stochastic General Equilibrium (DSGE)) modelle vir die analise van besigheidsiklus gebeure in die Suid Afrikaanse makroekonomie. Dit bestaan uit vier aparte dog onderling verwante dele wat in « logiese ontwikkeling vorm. Die eerste deel motiveer die gebruik van dié modelle en daarna word die teoretiese onderbou van die modelle daargestel. Die teoretiese onderbou word aangevul met gedetaileerde stappe van die a eiding van die verhoudings wat gebruik word om « model vir « klein oop ekonomie saam te stel. Die tweede deel oorweeg die eienskappe van Suid Afrikaanse makroekonomiese data wat relevant is vir « ekonometriese model in hierdie konteks. Dit sluit « bespreking in van die veranderlikes wat vir so « model gebruik kan word, asook « bespreking van die verskeie metodes wat gebruik kan word om die besigheidsiklus uit die data te identi seer. Die steekproefgrootte van die data word dan vasgestel, ná die moontlikheid van strukturele onderbrekings van tendens in die eerste en tweede momente van die data ondersoek is met behulp van verskeie enkel en meervoudige-veranderlike tegnieke. Die laaste hoofstuk van dié deel is « studie van verskeie maatstawwe van kern in asie (core in ation), waar « vergelyking getref word tussen die resultate van die volgende metodes toegepas op Suid Afrikaanse data: afgesnede gemiddeldes (trimmed means), dinamiese faktor modelle en verskeie golfvormige onderverdelings (wavelet decompositions). Die derde deel gebruik die datastel, wat in deel twee ontwikkel is, in die passing van « DSGE model wat die tipiese eienskappe van « klein oop ekonomie inkorporeer. Dit sluit « bespreking in van die rol van die wisselkoers in hierdie tipe modelle, en daar word empiries bevind dat die wisselkoers belangrike inligting bevat. Hierdie deel sluit ook « ondersoek in van optimale beleid in terme van die reaksie funksie van die sentrale bank. Die laaste deel van die proefskrif bestudeer die resultate van onlangse ontwikkellinge in DSGE modelle wat toegepas word op die Suid Afrikaanse ekonomie. Dit sluit « voorbeeld van « nie-liniêre model wat met behulp van « partikel lter (particle lter) geskat word en gebruik word vir vooruitskattings. Die vooruitskattings uit beide die liniêre en nie-liniêre modelle word dan vergelyk met dié verkry uit verskeie Vektor
Description
Thesis (PhD)--Stellenbosch University, 2014.
Keywords
UCTD, Dissertations -- Economics, Theses -- Economics, Business cycles -- South Africa, Equilibrium (Economics), Business cycles -- South Africa -- Econometric models
Citation