Three Applications of Time-Varying Parameter and Stochastic Volatility Models to the Malaysian and Australian Economy

Abstract

After the introductory chapter, this thesis comprises of three chapters that examines the application of time-varying parameter and stochastic volatility models to the Malaysian and Australian economy.

Chapter 2 aims to determine whether the propagation and transmission mechanism of Malaysian monetary policy differed during the Asian Financial Crisis of 1997/98 and the Global Financial Crisis of 2007/08. The methodology employs a time-varying vector-autoregression framework. The primary result is that despite having no evidence of time-variation in the propagation mechanism of Malaysian monetary policy the average contribution of a monetary policy shock to the variability of each macroeconomic variable-Real GDP, Inflation and the Nominal Effective Exchange Rate-differs between the two crises. This finding suggests that despite the propagation mechanism being relatively constant, Malaysia's monetary policy transmission mechanism evolves over time. We believe that the main mechanism driving this evolution is the time-variation in the variance-covariance matrix of the shocks of the model, not the coefficients. We also find some evidence that the implementation of capital controls reduced the influenceability of monetary policy on the Malaysian economy.

Chapter 3 investigates whether incorporating time variation and fat-tails into a suite of popular univariate and multivariate Gaussian distributed models can improve the forecast performance of key Australian macroeconomic variables: real GDP growth, CPI inflation and a short-term interest rate. The forecast period is from 1992Q1 to 2014Q4, thus replicating the central banks forecasting responsibilities since adopting inflation targeting. We show that time varying parameters and stochastic volatility with Student's-t error distribution are important modeling features of the data. More specifically, a vector autoregression with the proposed features provides the best interest and inflation forecasts over the entire sample. Remarkably, the full sample results show that a simple rolling window autoregressive model with Student's-t errors provides the most accurate GDP forecasts.

Chapter 4 estimates a time-varying parameter Panel Bayesian vector autoregression with a new feature: a common stochastic volatility factor in the error structure, to assess the synchronicity and the nature of Australian State business cycles. The common stochastic volatility factor reveals that macroeconomic volatility or uncertainty was more pronounced during the Asian Financial Crisis as compared to the more recent Global Financial Crisis. Next, the Panel VAR’s common, regional and variable specific indicators capture several interesting economic facts. In the first instance, the fluctuations of the common indicator closely follow the trend line of the Organisation for Economic Co-operation and Development composite leading indicators for Australia making it a good proxy for nationwide business cycle fluctuations. Next, despite significant co-movements of Australian States and Territory business cycles during times of economic contractions, the regional indicators suggest that the average degree of synchronisation across the Australian States and Territories cycles in the 2000s is only half of that presented in the 1990s. Given that aggregate macroeconomic activity is determined by cumulative activity of each of the nation states, the results suggests that the Federal Government should award state governments greater autonomy in handling state specific cyclical fluctuations.