Groves, Jeremy R.
Campbell III, Carl M.
Ph.D. (Doctor of Philosophy)
Department of Economics
The doctoral dissertation entitled Three Essays in Applied Econometrics is based on the application of modern techniques of applied econometrics in order to infer main interrelations and causalities in various fields of quantitative economics. It includes an eclectic mix of both frequentist and Bayesian approaches, with the main goal to quantify and estimate a set of variables that significantly determine certain economic outcomes, in this case NBA efficiency scores, per capita growth rates, and Covid-19-induced risk exposure, market volatility, and cross-market integration. These topics are, at the sametime, at the core of the research interest in the following three independent chapters.
The first chapter employs a slack-based Data Envelopment Analysis (DEA) model to estimate efficiencyscores of National Basketball Association (NBA) teams (and cohorts). A second-stage regression analysis predicts the effects that experience and competitiveness have on in-game efficiency over the period 2005- 2018. My results indicate that selected "big market teams" (e.g., Cleveland, Golden State, Los Angeles, Miami) are actually the least efficient, while certain "small market" teams (Denver, Houston, Indiana, Minnesota, Phoenix, Oklahoma) consistently perform at near-optimum efficiency levels. The estimated efficiency scores tend to exhibit dispersion levels that are downward-trending, implying that the NBA is becoming slightly more competitive over time. The regression results reveal a negative relationship be- tween efficiency and team experience, due to decreased athletic ability and diminishing marginal utility of wealth. Finally, it has been shown that measures of team competitiveness and public support are not statistically significant for in-game efficiency.
The second chapter investigates the underlying determinants of economic growth in 45 advanced economiesthrough the application of cross-sectional Bayesian Model Averaging (BMA). Data set consists of some 29 macroeconomic, demographic, and institutional variables over the period 1996-2017. This research is aimed to address the problem of model uncertainty in cross-sectional growth regressions, and to prescribe a universal policy recipe for faster economic growth in the long-run. The most robust determinants of per-capita GDP growth rates, measured by the posterior inclusion probability (PIP) are life expectancy, the initial GDP growth rate, the level of trade openness (as proxied by exports), education index, and remittance outflows. This conclusion is robust to different model and parameter (g-prior) specifications. In addition, the results reveal a significant non-linear relationship between several of these variables; however, there is a significant imprecision in the model-generated forecasts (Switzerland and Malaysia excepted), meaning that the BMA exhibits lukewarm out-of-sample performance.
The closing chapter of the thesis deals with the Covid-19-induced effects on the global financial markets (Jan 2020 - June 2020), including market reactions to the crisis-related policy measures such as near- zero interest rate and fiscal stimulus payment (The CARES Act). The study implements Conditional Value-at-Risk (CVaR), MCMC stochastic volatility, and Copula bivariate tail dependence to estimate risk-exposure and cross-market integration amid the pandemic , precisely in the pre-pandemic, early- pandemic, and pandemic periods. The results indicate that risk-exposure the stock markets has exploded in both early pandemic and pandemic periods; on the other hand, the crypto market experienced a decline in risk exposure in the early-pandemic period, but a rapid risk-exposure increase during the pandemictime. The effects of the pandemic on the commodity market are asset-specific. The anti-crisis measures cooled down the markets initially (except in the case of Brent market), but with no long-lasting effects. The Covid-19-induced crisis made the markets increasingly cross-dependent, which calls on a strong macroprudential regulatory agenda, aimed at reducing the probability of a global-scale financial contagion to occur.
Davidovic, Milivoje, "Three Essays in Applied Econometrics" (2021). Graduate Research Theses & Dissertations. 6964.
Northern Illinois University
Rights Statement 2
NIU theses are protected by copyright. They may be viewed from Huskie Commons for any purpose, but reproduction or distribution in any format is prohibited without the written permission of the authors.