Browsing by Subject "Panel data"
Now showing 1 - 7 of 7
Results Per Page
Sort Options
Item Does trade cause inequality(2015-05) Zhang, Haoran; Shively, Thomas S.; Gawande, KishoreThe relationship between international trade and income distribution of countries becomes a hot topic in economics research. This paper use random forest method and stepwise regression method to complete variables selection work from a big panel data set with many economic variables. Analysis of an unbalanced panel of country level data reveals that the trade will reduce income inequality in most situations. The coefficients for trade variables are significant in both two types of models, i.e., with and without considering about country effects. But when we split data set into two groups, the coefficients are significant for developed countries but not significant for developing countries.Item The life insurer Risk-Based Capital ratio : panel data analysis(2013-05) Beisenov, Aidyn; Sager, Thomas W.Many studies suggest the ability of the NAIC Risk-Based Capital ratio (RBC ratio) to predict insurer insolvency. Based on the US life insurer (insurer) data for the period of 2005 to 2008, this study finds explanatory variables that have a statistically significant relationship with the RBC ratio. Advantages of panel data over cross-sectional and time series data analysis are exploited to make valid inference on coefficients of the explanatory variables. Testing for unobserved insurer and time effects and for dependence between these effects and the explanatory variables indicates the appropriateness of the fixed insurer and time effects model. Based on the ordinary least squares estimates, it is found that insurers' size, capital-to-asset ratio, and return on capital have a statistically significant relationship with the RBC ratio. Additionally, health product, annuity product, opportunity, and regulatory risks of insurers are related to the RBC ratio. Accounting for heteroscedasticity and autocorrelation for a given insurer yields the same coefficient estimates, but increased standard errors.Item Limited Dependent Variable Correlated Random Coefficient Panel Data Models(2012-10-19) Liang, ZhongwenIn this dissertation, I consider linear, binary response correlated random coefficient (CRC) panel data models and a truncated CRC panel data model which are frequently used in economic analysis. I focus on the nonparametric identification and estimation of panel data models under unobserved heterogeneity which is captured by random coefficients and when these random coefficients are correlated with regressors. For the analysis of linear CRC models, I give the identification conditions for the average slopes of a linear CRC model with a general nonparametric correlation between regressors and random coefficients. I construct a sqrt(n) consistent estimator for the average slopes via varying coefficient regression. The identification of binary response panel data models with unobserved heterogeneity is difficult. I base identification conditions and estimation on the framework of the model with a special regressor, which is a major approach proposed by Lewbel (1998, 2000) to solve the heterogeneity and endogeneity problem in the binary response models. With the help of the additional information on the special regressor, I can transfer a binary response CRC model to a linear moment relation. I also construct a semiparametric estimator for the average slopes and derive the sqrt(n)-normality result. For the truncated CRC panel data model, I obtain the identification and estimation results based on the special regressor method which is used in Khan and Lewbel (2007). I construct a sqrt(n) consistent estimator for the population mean of the random coefficient. I also derive the asymptotic distribution of my estimator. Simulations are given to show the finite sample advantage of my estimators. Further, I use a linear CRC panel data model to reexamine the return from job training. The results show that my estimation method really makes a difference, and the estimated return of training by my method is 7 times as much as the one estimated without considering the correlation between the covariates and random coefficients. It shows that on average the rate of return of job training is 3.16% per 60 hours training.Item Multiple regression applications to capital structure modeling for life insurers(2009-12) Kanwar, Ridhi; Sager, Thomas W.; Powers, Daniel A.Like any other company, life insurance companies maintain a combination of debt and equity for their long-term financing, which forms their capital structure. Many theories have been developed in the literature to focus upon the determinants that are likely to affect leverage decisions of the life insurance firms in the post life Risk-Based Capital (RBC) regulation era. This report documents the application of multiple regression techniques to derive and analyze a Capital Asset Ratio (CAP) model based on the data pertaining to a large number of life insurance companies during 2000 to 2004. The data set is organized as panel data. Model coefficients, together with the error structure, are analyzed using SAS software to develop a valid model that tries to explain the Capital to Asset Ratio (CAP) for life insurers in terms of various variables of interest. The latter include return on capital, total assets, and two measures of risk: asset risk and product risk, etc. A balanced panel dataset was extracted from the given unbalanced input dataset containing missing entries. In addition, a selected few of the explanatory variables were chosen from a large group present in the input dataset based on previous work on relations among asset risk, product risk and capital in the life insurance industry by Etti G. Baranoff and Thomas W. Sager (2002). Fixed Effects model was chosen based on the assumption that the firm-specific effects were correlated to the explanatory variables. Differencing method was employed so that OLS estimator could safely be used for the coefficients in the regression model. Based on the proposed model, it is found that Capital to Asset Ratio has positive relationships with product risk and return on capital, with the corporate form of organization, and with membership in an affiliated group of companies. On the other hand, it has a negative relationship with company’s size and the ratio of life premiums or annuity premiums to the total premiums generated.Item A new estimation approach for modeling activity-travel behavior : applications of the composite marginal likelihood approach in modeling multidimensional choices(2011-08) Ferdous, Nazneen; Bhat, Chandra R. (Chandrasekhar R.), 1964-; Machemehl, Randy; Abrevaya, Jason; Waller, Steven; Stolp, ChandlerThe research in the field of travel demand modeling is driven by the need to understand individuals’ behavior in the context of travel-related decisions as accurately as possible. In this regard, the activity-based approach to modeling travel demand has received substantial attention in the past decade, both in the research arena as well as in practice. At the same time, recent efforts have been focused on more fully realizing the potential of activity-based models by explicitly recognizing the multi-dimensional nature of activity-travel decisions. However, as more behavioral elements/dimensions are added, the dimensionality of the model systems tends to explode, making the estimation of such models all but infeasible using traditional inference methods. As a result, analysts and practitioners often trade-off between recognizing attributes that will make a model behaviorally more representative (from a theoretical viewpoint) and being able to estimate/implement a model (from a practical viewpoint). An alternative approach to deal with the estimation complications arising from multi-dimensional choice situations is the technique of composite marginal likelihood (CML). This is an estimation technique that is gaining substantial attention in the statistics field, though there has been relatively little coverage of this method in transportation and other fields. The CML approach is a conceptually and pedagogically simpler simulation-free procedure (relative to traditional approaches that employ simulation techniques), and has the advantage of reproducibility of the results. Under the usual regularity assumptions, the CML estimator is consistent, unbiased, and asymptotically normally distributed. The discussion above indicates that the CML approach has the potential to contribute in the area of travel demand modeling in a significant way. For example, the approach can be used to develop conceptually and behaviorally more appealing models to examine individuals’ travel decisions in a joint framework. The overarching goal of the current research work is to demonstrate the applicability of the CML approach in the area of activity-travel demand modeling and to highlight the enhanced features of the choice models estimated using the CML approach. The goal of the dissertation is achieved in three steps as follows: (1) by evaluating the performance of the CML approach in multivariate situations, (2) by developing multidimensional choice models using the CML approach, and (3) by demonstrating applications of the multidimensional choice models developed in the current dissertation.Item Three essays in empirical industrial organization(2006-12) Dunn, Abraham C.; Hendricks, KennethThere are many differentiated product industries in which firms offer multiple products in the same market. In making strategic decisions regarding entry, quality and quantity to be supplied for their multiple products firms must consider the competition with rivals as well as cannibalization of their own products that are close substitutes. In this setting, understanding the relationship between the behavior of consumer demand and firms decisions' regarding product characteristics and strategic variables like advertising are fundamental issues in industrial organization. This dissertation empirically explores these fundamental issues in the pharmaceutical and airline industries. The first paper of my dissertation estimates consumer demand for different anti-cholesterol drugs using panel data on a nationally representative sample of individuals who were diagnosed with cholesterol problems in the period 1996-2002. The data provides detailed information on individuals' medical conditions, medical and drug insurance coverage, drug purchases (if any), and other demographic and medical information. Individuals choose whether to purchase an anti-cholesterol rug and, if so, which drug to buy. The model permits flexible substitution patterns among drug choices and persistence in those choices by incorporating both observed and unobserved consumer heterogeneity. The estimates suggest that lower income patients without prescription drug insurance are very price sensitive: they are less likely to use drugs and, if they do use them, they tend to purchase the less expensive drugs. I find that roughly 500 thousand individuals without drug insurance who are currently not purchasing anti-cholesterol drugs would do so in the counterfactual world in which they are given the standard co-payment plan. The second paper also looks at consumer demands for anti-cholesterol drugs. While the first paper focused on the differentiated products, this paper explores the market expansion effects of direct-to-consumer advertising (DTCA). The study combines the individual data used in the first paper with monthly expenditure data on DTCA for the period 1996-2002. The dynamic demand model estimated in this paper explores the heterogeneous effects of DTCA. Overall, I find a positive effect from DTCA with short term elasticity of 0.107. Through persistence in consumer demand this effect lasts over multiple time periods. I find that individuals not taking a cholesterol drug respond more to advertising than those on the drug. In addition, I find that less educated individuals, those that may be unaware of their health condition, and those without health insurance are most responsive to DTCA. Finally, the third paper studies the effect of product ownership and quality on entry in the airline industry. Specifically, this paper empirically examines the decision of an airline to offer high quality nonstop service between cities given that the airline may or may not be offering lower quality one-stop service. I find that airlines that offer one-stop service through a hub are less likely to enter that same market with nonstop service than those that do not. In addition, the quality of the one-stop service is another determinant of entry. Airlines are more likely to enter a market with nonstop service if their own or their rival's one-stop service in the market are of lower quality.Item Wastewater expenditure effects on in-stream bacteria pollution in the Rio Grande / Río Bravo post-NAFTA : evidence from panel data estimations(2014-08) Torres, Adam Jared; Stolp, Chandler; Olmstead, Sheila M.The United States and Mexico share responsibility in preserving the quality of their international river system, the Rio Grande / Río Bravo, and several international treaties govern the quantity of water each country must give and take. Because no treaty establishes joint standards for the quality of the river, the North American Agreement on Environmental Cooperation (NAAEC) was created in 1993 as a declaration of principles and objectives concerning the conservation and the protection of the environment as well as a guide of concrete measures to further cooperate on these matters. One particular goal of the NAAEC was to improve water quality in the US-Mexico Border Region, ensuring a clean, safe, and reliable water supply for the area. Although the US and Mexican federal governments have made substantial technical and financial commitments through binational agencies like the North American Development Bank (NADB) and the Border Environment Cooperation Commission (BECC), few empirical studies have assessed the impact of binational expenditures on wastewater infrastructure in this region. This report uses longitudinal panel data regression models to estimate the impact of capital expenditures on water quality made by binational, federal, and state water quality management institutions from 1995 to 2012. This analysis considers expenditures made on both sides of the Rio Grande watershed that constitutes the international border, beginning with El Paso, Texas and ending in the Gulf of Mexico.