Alumni Dissertations and Theses

 
 

Alumni Dissertations and Theses

Filter Dissertations and Theses By:

 
 
  • ESSAYS ON ROBUST ESTIMATORS FOR NON-IDENTICALLY DISTRIBUTED OBSERVATIONS IN SPATIAL ECONOMETRIC AND TIME SERIES MODELS

    Author:
    Suleyman Taspinar
    Year of Dissertation:
    2014
    Program:
    Economics
    Advisor:
    Wim Vijverberg
    Abstract:

    This thesis proposal consists of three essays on the estimation methods and applications of spatial econometric models and one essay on the generalized autoregressive conditionally heteroskedastic (GARCH)-type models in financial time series. The first essay discusses the heteroskedasticity robust generalized method of moments estimator (RGMME) for the spatial models that allow for spatial dependence in both the dependent variable and the disturbance term (SARAR(1,1)). First, we show that the maximum likelihood estimator (MLE) is generally inconsistent in the presence of unknown heteroskedasticity. Then, we extend robust GMM approach in Lin and Lee (2010) to SARAR(1,1). The large sample properties are rigorously studied and presented for the RGMME. Through a comprehensive Monte Carlo study, we compare the finite sample properties of the RGMME with some other estimators proposed in the literature. The second essay focuses on the GMM estimation of the spatial autoregressive models which impose a moving average process for the disturbance term (SARMA). We extend the best GMM estimator (BGMME) of Liu et al. (2010) to the SARMA models and provide the best set of instruments for the SARMA(1,1) and the SARMA(0,1) specifications. The large sample properties are rigorously studied and presented for the BGMME. The finite sample properties are investigated through an extensive Monte Carlo study. To confirm our results from the Monte Carlo study, we replicate the results for the SARMA(1,1) specification in Behrens et al. (2012) in an empirical illustration. The third essay investigates the effect of foreign direct investment (FDI) on economic growth through a spatially augmented Solow growth model. The current literature on the relationship between FDI and economic growth uses canonical cross-country growth regression specifications that are derived from the textbook Solow growth model for closed economies. We claim that these specifications cannot reflect the relationship between economic growth and FDI, because they model each country as an isolated island that does not interact with the rest of the world. On the other hand, a spatially augmented Solow growth model allows for technological interdependence among countries through spatial externalities. The modified growth model yields regression specifications that properly account for spatial autocorrelations. We construct a panel of 85 countries for the period 1980-2010 and estimate the modified specifications with the tools from spatial econometrics. Our findings indicate that FDI inflows have a significant positive effect on the growth rate of host countries. The final essay proposes a flexible distribution for the maximum likelihood estimation of the GARCH-type time series models. The new distribution can better account for the potential skewness and leptokurticity in the driving noise sequence. We study the large sample properties of the new estimator following the methodology presented in Francq and Zakoïan (2004). To investigate the finite sample properties of the new estimator, we first conduct a Monte Carlo study. Furthermore, to test the relative out-of-sample predictive power of the new estimator, we test for its prediction power on two data sets using the methods described in White (2000) and Hansen et al. (2003).

  • Default Risk

    Author:
    Zeynep Topaloglu
    Year of Dissertation:
    2010
    Program:
    Economics
    Advisor:
    Thom Thurston
    Abstract:

    The hazard rate models used in recent bankruptcy literature assume censoring and default are two independent events, which means the censored company will eventually default. However we believe there will be a portion in the censored group that will be long-term survivors and we propose a mixture model of survivors and risky companies. Moreover this dissertation models the event and the timing of default incident at the same time. For the event of default and the timing of default we utilize a logistic regression. The results have justified the advantage of our model over the standard hazard rate models and proved its predictive power. The companies identified as high default risk by our model proved to deliver extremely low returns in the market.

  • EXCHANGE RATES, PRICES AND PROFITABILITY: THE TURKISH EVIDENCE

    Author:
    Nazli Toraganli
    Year of Dissertation:
    2012
    Program:
    Economics
    Advisor:
    JOHN DEVEREUX
    Abstract:

    This dissertation consists of three essays. The first essay, published in the Turkish Central Bank Review I, 2010, examines the impact of exchange rate movements on export prices and profitability, at sectoral and at firm level respectively, for Turkish manufacturing. The results suggest that i) Turkish firms tend to stabilize their local currency prices implying that they have market power in international markets, and ii) exchange rate movements affect the profitability of firms. The results also show that the profits of export oriented firms are more likely to be affected by exchange rate movements. In the second essay, I develop a two-country model of international trade, built on Melitz (2003) model, where plant level heterogeneity is the key assumption. In this framework I analyze the impact of currency appreciation on productivity thresholds using a Newton Algorithm. The numerical simulation suggests that exchange rate appreciation increases productivity thresholds for both exporting and domestic production. In addition, I incorporate liquidity constraints in line with Manova (2010) into a model of heterogeneous firms and numerically show that, in the presence of liquidity constraints, the productivity threshold for exporting increases. The third essay, co-authored with Professor Yazgan from Bilgi University, uses unique firm level dataset, to test the impact of currency appreciation on the survival behavior and sales of Turkish firms in traded and nontraded industries for 2002-2009. The results suggest that real exchange rate appreciation decreases the probability of survival in both traded and nontraded industries. We find that high (low) productivity firms have higher (lower) probability of survival than low (high) productivity firms in both traded and nontraded industries as a result of domestic exchange rate appreciation. We further test the impact of currency appreciations on the sales of continuing firms belonging to both traded and nontraded industries. Our results show that real exchange rate appreciations do not have any impact on continuing firms' sales in both traded and nontraded industries.

  • Robust Estimators for Finite Mixtures of Count Data Regression Models and their Applications

    Author:
    Ti-Jen Tsao
    Year of Dissertation:
    2010
    Program:
    Economics
    Advisor:
    Partha Deb
    Abstract:

    Finite mixtures of count data regression models have been successfully used for modeling discrete responses arising from heterogeneous populations. But the maximum likelihood (ML) estimator for such models are sensitive to data contamination and extreme values. This dissertation develops two robust estimators for finite mixtures of count data regression models. One is the minimum Hellinger distance (MHD) estimator and the other is the minimum L2 error (L2E) estimator, a special case of the minimum density power divergence estimator. Two Monte Carlo simulation studies show that the MHD and L2E estimators are more robust than the ML one but come at the cost of efficiency. However, the robustness property of the MHD and L2E estimators is deteriorated as the mixing probability approaches one. For empirical application, this dissertation uses the data from Dionne et al. (1996), the extent of non-payments of personal loans in Spain, and from Deb and Trivedi (2002), counts of utilization from the RAND Health Insurance Experiment, respectively. The estimated results show that the two-component Poisson mixture regression model is the best fit model for the first data set and the two-component negative binomial one mixture regression model for the second data set. But both of the model specifications are preferred to be estimated by the ML estimation that could be attributed to the flexibility of the finite mixture model and data processing procedures.

  • Essays in corporate finance

    Author:
    Milos Vulanovic
    Year of Dissertation:
    2010
    Program:
    Economics
    Advisor:
    Armen Hovakimian
    Abstract:

    ABSTRACT ESSAYS IN CORPORATE FINANCE by Milos Vulanovic Adviser: Professor Armen Hovakimian This dissertation consists of two essays on corporate finance. In the first essay we test the pecking order theory by examining how firms finance maturing long-term debt. This allows us to accomplish three goals: resolve the issues of debt capacity and the endogeneity of financing deficit; examine the role of internal financing; and generate evidence regarding the order in which different sources of financing are used. We determine that firms use internal funds before they issue new debt to refinance maturing long-term debt. Firms with more cash on hand are less likely to issue new debt to refinance. On average, each marginal dollar of maturing long-term debt is fully financed with the issuance of new debt. In the second essay, we study characteristics of Specified Purpose Acquisition Companies (SPACs) and examine the performance of their securities over time. We find that SPACs represent a fairly unique way to raise capital, The incentives of their founders, underwriters, and investors are interdependent and successful business combinations generally result in significant returns to founders. We also show that different SPAC securities generate different reactions in response to the announcement news regarding their corporate status. While holders of all three securities realize abnormal returns on the announcement day, the strongest reaction is observed among the investors holding warrants, while common stock holders tend to react very mildly.

  • Three Essays in Health Economics

    Author:
    Xu Wang
    Year of Dissertation:
    2012
    Program:
    Economics
    Advisor:
    Michael Grossman
    Abstract:

    This dissertation consists of three essays. In my first essay, I examine the relationship of the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC program) to breastfeeding. Although WIC promotes breastfeeding among its participants through education, counseling, and the provision of enhanced food packages to breastfeeding women, the program has been criticized for discouraging breastfeeding by providing free infant formula. In order to estimate the extent to which participation in WIC discourages breastfeeding, I employ a methodology that disentangles selection bias associated with WIC participation from the incentives associated with the provision of free infant formula. Findings suggest that postpartum entrants are less likely to breastfed for at least 6 months and have shorter breastfeeding durations than non-participants, and the effects are significantly larger among twin mothers than among singleton mothers. In my second essay, I investigate the association between WIC participation and infant health. How effective WIC is at improving birth outcomes is under debate. Identifying treatment effect is challenged by selection bias and gestational age bias. We use twins to minimize selection bias associated with WIC participation because twin pregnancy increases the probability of adverse birth outcomes significantly but is unlikely related to other risky behaviors. Our focus is on measures of fetal growth as outcomes amenable to nutritional supplementation. Our findings from two national datasets, PNSS and ECLS-B, suggest that prenatal WIC participation has very limited effect on fetal growth. We do not find evidence of causal effect between WIC and better birth outcomes, especially among twin births. In my third essay, I turn my interest to a different research question, the association between retirement and alcohol consumption. Retirement is life transition whose significance may provoke lifestyle and health behavioral alterations such as alcohol consumption. We examine the effect of retirement on subsequent period alcohol consumption within a two period follow up. We use seven waves of the data from Health and Retirement Study (HRS) and found retirement lead to consume 1.3 more alcoholic drinks per day within men. No effect has been found within retired women.

  • Essays on the Determinants and Consequences of Physical Activity

    Author:
    Jin Wang
    Year of Dissertation:
    2013
    Program:
    Economics
    Advisor:
    Michael Grossman
    Abstract:

    The first essay focuses on the economic determinants of physical activity. Physical activity can be decomposed by intensity and duration. The intensity of physical activity is defined by total energy consumed on all physical activities divided by total hours. The intensity of an activity is measured by its metabolic equivalent (MET) value per hour. One MET is defined as the energy expended to lie or sit quietly. One novel aspect of the study is an investigation of the hypothesis that, because their time is more valuable, higher wage individuals may choose to exercise more intensively but for shorter durations. That hypothesis and others are explored in the context of fully specified demand functions for the duration and intensity of activities performed in the market and nonmarket sectors. The instrumental variable method is adopted to estimate the demand functions in which the endogenous variable leisure physical activity is included. Econometric issues related to omitted variable bias and selection bias are addressed as part of the empirical investigation. The second essay investigates to what extent physical activity influences health status. To study the heterogeneous effects of physical activity on dichotomous health outcome for three groups of people, the analyses of three stratifications are conducted for education, occupation, and physical activity intensity. Probit (ordered probit) models are utilized to address the research question. I find that for less educated or white collar workers, a rise in average physical intensity increases the probability of reporting very good or excellent health, while longer time spent on physical activity may benefit health for highly educated workers.

  • Credit Risk Pricing in Single Name Corporate CDS

    Author:
    Yikai Wang
    Year of Dissertation:
    2010
    Program:
    Economics
    Advisor:
    Liuren Wu
    Abstract:

    The paper examines the determinants of the dynamics and term structure of credit default swap (CDS) spreads. I focus on roles of the interest rate risk and default risk. I extract interest rate factors from libor-swap curve, based on affine models. With a large data set, I test a three-factor CDS spread term structure model for 100 companies, using exact matching technique. The factors consist of two interest rate factors and one default risk factor. The estimation shows intricate interaction between interest rate factor and credit risk factor in determining the default process. I find that positive shock of the first interest rate factor will increase subsequent default risk factor, while positive shock on the second interest rate factor will decrease the subsequent default risk factor. In terms of direct effect of interest rate factors on the default process, positive shock on the first interest rate factor will decrease the default arrival rate at the moment, while positive shock on the second interest rate factor will increase the default arrival rate at the moment. Most of the effects of the shocks to interest rate factors will be sent to the default process indirectly through their impact on the subsequent move of the credit risk factor. CDS spreads of high credit rating companies are less responsive to the shocks on default risk than those of low credit rating companies. The work enhances our understanding of the process which underlies the CDS spreads movement and regimes. Such knowledge is essential for CDS pricing, risk measurement and management, hedging the related risk.

  • Essays in Unauthorized Immigration and Migration

    Author:
    Bilesha Weeraratne
    Year of Dissertation:
    2012
    Program:
    Economics
    Advisor:
    David Jaeger
    Abstract:

    This dissertation consists of three essays. The first essay develops a new methodology defined as the microdata-based methodology to identify unauthorized immigrants in the US. This identification is based on a discrete choice model on observed authorized and unauthorized immigrant data in 1986 with adjustments for time dynamics between 1986 and 2010 and for calibration in new data. This method produces an algorithm to identify unauthorized immigrants in the American Community Survey (ACS) 2010, and estimates that there were 7,700,869 adult unauthorized immigrants in the US on January 1, 2011. The essay includes detailed descriptive analyses of unauthorized immigrants and their children. In the second essay, I evaluate the impact of parents' legal status on the high school drop out probability of 16-18 year old children. The analysis uses data from the ACS 2010, and the methodology is a linear probability model. The study finds that when controlling for other covariates, a child of unauthorized immigrants has a lower drop out probability than a similar child of authorized immigrant parents, and among unauthorized immigrant parents, a non-citizen child has a higher drop out probability than a similar native-born child. In mixed families, among unauthorized immigrant mothers, a non-citizen child has a higher drop out probability than an otherwise similar native-born child, and children with the parent combination of highly unauthorized and highly legal immigrants have a higher high school drop out probability than those with highly unauthorized and unsure legal immigrant parents. In the third essay, I estimate a multinomial logit model to analyze the contextual determinants of labor migration in Sri Lanka, using data from the Consumer Finance and Socio Economic Survey 2003/4 of the Central Bank of Sri Lanka. The study finds higher internal migration probabilities for residents of rural areas, districts with a lower degree of structural transformation, and districts with a larger share of population in 19-34 years. Internal migration probabilities are lower for residents of districts with large labor forces.

  • Monetary Policy Shocks and Their Effects on Business Borrowings, Aggregate Output, and Prices

    Author:
    Yao Yao
    Year of Dissertation:
    2012
    Program:
    Economics
    Advisor:
    Thom Thurston
    Abstract:

    Developed from previous financial acceleration research, this thesis employs two VAR (vector-auto regression) systems to measure monetary policy shocks and their effects. The first uses the conventional Cholesky method and the second uses the methods developed by Bernanke and Blinder (1992) and Bernanke and Mihov (BM 1998). I use the federal funds rate and as the two monetary policy instruments in the VAR systems. With the Cholesky method, net financial borrowings rise (fall) for about one year after contractionary (easy) monetary policy shocks. This is consistent with what has been found by Christiano, Eichenbaum, and Evans (CEE 1996). The responses of corporate sector appear to explain the entire business sector, with no significant results found from the impulse responses of the other two sectors (noncorporate and farm). With the BM model, easing policy shocks lead to an initial fluctuation within the first half of year and the prolonged decrease in borrowings after half of year. The responses of borrowings in the BM model demonstrate different amplitudes and patterns from the Cholesky method following the policy shocks, this is from the structural shocks are defined differently in these two VARs. However, the decreases (rise) in borrowings after the easy (tight) monetary shocks are the same. In both systems, there is a significant expansion of real GDP following monetary easing. The price puzzle, that price levels are positively related to monetary tightening, does not appear in the plots of impulse response functions with the Cholesky method but they appear in some versions of the BM model.