Explorations in Simulation Based Techniques

My dissertation is composed of three chapters and although the chapters are different in terms of topics and methods, they all utilize simulation based techniques, hence the title “Explorations in Simulation Based Techniques”. The first chapter focuses on estimation methods for quantile regression, the second chapter presents the Bayesian approach to quantile regression in ordinal models and the third chapter examines the deregulatory effect on the transmission of monetary policy to U.S. residential investment.;The first chapter demonstrates that metaheuristic algorithms can provide a useful general framework for estimating both linear and nonlinear econometric models. Two metaheuristic algorithms—firefly and accelerated particle swarm optimization—are employed in the context of several quantile regression models. The algorithms are stable and robust to the choice of starting values and the presence of various complications (e.g. non-differentiability, parameter restrictions, discontinuity, possible multimodality, etc.). Two comparative studies involving an autoregressive model and a conditional scale autoregressive conditional heteroscedasticity model, demonstrate the performance of metaheuristic algorithms relative to existing approaches. In addition to these examples, the paper also offers an application to consumption behavior in which the presence of constraints makes existing techniques difficult to implement, but metaheuristic algorithms are straightforward to apply. The findings indicate that, contrary to popular perception, marginal propensity to consume is highest in Quarter 3 for each of the sample years. Moreover, pre- and post-recession comparisons reveal interesting asymmetries in consumption behavior.;The second chapter introduces a Bayesian estimation method for quantile regression in univariate ordinal models. Two algorithms are presented that exploit the latent variable inferential framework of Albert and Chib (1993), capitalize on the normal-exponential representation of the asymmetric Laplace (AL) distribution, and judiciously select scale restriction to simplify the sampling procedure. Estimation utilizes Markov chain Monte Carlo (MCMC) simulation methods—either Gibbs sampling together with the Metropolis-Hastings (MH) algorithm or only Gibbs sampling. The algorithms are demonstrated in two simulation studies and employed to analyze problems in economics (educational attainment) and political economy (public opinion on a recently proposed tax policy).;The last chapter looks for structural changes in the transmission of monetary policy to residential investment following two deregulations: the abolition of Regulation Q and the ratification of the Gramm-Leach-Bliley Act (GLBA). The potential changes in transmission are studied within a time varying vector auto-regression (TVAR) model that incorporates stochastic volatility. Results suggest that the abolition of Regulation Q has almost no change in the response of residential investment to a monetary policy shock. The period (1985–1999) following the abolition is characterized by smaller variance of shocks to residential investment. After the ratification of GLBA, residential investment is more responsive to monetary policy and there is a sudden increase in the variance of shocks to residential investment and house price. The increase in the variance of shocks is consistent with the view that GLBA was responsible, amongst other factors, for soaring home prices and the sub-prime mortgage crisis. On the modeling aspect, use of the stochastic volatility affects the size of impulse responses and is necessary to reflect changes in volatility overtime.