Browsing by Author "Howard Bondell, Committee Member"
Now showing 1 - 11 of 11
- Results Per Page
- Sort Options
- Benefits, Costs and Carbon Abatement of Building Energy Efficiency Standards in China.(2010-04-30) Qi, Ji; Michael Roberts, Committee Chair; Roger von Haefen, Committee Member; Howard Bondell, Committee Member
- Boosting methods for variable selection in high dimensional sparse models(2009-08-27) Hwang, Wook Yeon; Hao Helen Zhang, Committee Member; Howard Bondell, Committee Member; Wenbin Lu, Committee Member; Subhashis Ghosal, Committee ChairFirstly, we propose new variable selection techniques for regression in high dimensional linear models based on a forward selection version of the LASSO, adaptive LASSO or elastic net, respectively to be called as forward iterative regression and shrinkage technique (FIRST), adaptive FIRST and elastic FIRST. These methods seem to work better for an extremely sparse high dimensional linear regression model. We exploit the fact that the LASSO, adaptive LASSO and elastic net have closed form solutions when the predictor is one-dimensional. The explicit formula is then repeatedly used in an iterative fashion until convergence occurs. By carefully considering the relationship between estimators at successive stages, we develop fast algorithms to compute our estimators. The performance of our new estimators is compared with commonly used estimators in terms of predictive accuracy and errors in variable selection. It is observed that our approach has better prediction performance for highly sparse high dimensional linear regression models. Secondly, we propose a new variable selection technique for binary classification in high dimensional models based on a forward selection version of the Squared Support Vector Machines or one-norm Support Vector Machines, to be called as forward iterative selection and classification algorithm (FISCAL). This methods seem to work better for a highly sparse high dimensional binary classification model. We suggest the squared support vector machines using 1-norm and 2-norm simultaneously. The squared support vector machines are convex and differentiable except at zero when the predictor is one-dimensional. Then an iterative forward selection approach is applied along with the squared support vector machines until a stopping rule is satisfied. Also, we develop a recursive algorithm for the FISCAL to save computational burdens. We apply the processes to the original onenorm Support Vector Machines. We compare the FISCAL with other widely used binary classification approaches with regard to prediction performance and selection accuracy. The FISCAL shows competitive prediction performance for highly sparse high dimensional binary classification models.
- Changes in the Compensation Package of the Board of Directors and CEOs after the Corporate Scandal in late 2001.(2010-12-06) Chen, Yuanyuan; Charles Knoeber, Committee Chair; Mark Walker, Committee Member; Howard Bondell, Committee Member
- Evaluating the Effectiveness of Pre-Retirement Seminars: A Case Study of Progress Energy(2010-04-30) Guo, Qi; Melinda Morrill, Committee Chair; Robert Clark, Committee Co-Chair; Howard Bondell, Committee MemberGUO, QI. Evaluating the Effectiveness of Pre-Retirement Seminars: A Case Study of Progress Energy. (Under the direction of Melinda Sandler Morrill and Robert Clark.) This research is based on the data collected from the pre-retirement seminars of Progress Energy during 2008 and 2009. This research verifies the fact that the current employees are lacking of certain knowledge of retirement policies and plans, and the highly evaluated pre-retirement seminars provided by PGN did improve the necessary financial literacy of the attendees. The findings also indicate that the effectiveness of the seminars changed between 2008 and 2009 when the socio-economic and demographic characteristics of the attendees were stable over time. A further discussion on the possible impacts emphasizes on both the external changes brought by economic downturn and the internal changes of the seminars in 2009.
- Fast FSR Methods for Second-Order Linear Regression Models(2008-08-11) Crews, Hugh Bates; Howard Bondell, Committee Member; Jason Osborne, Committee Member; Dennis Boos, Committee Co-Chair; Leonard Stefanski, Committee Co-Chair
- Funding Game and Nongame Conservation: An Analysis of Willingness to Pay for Conservation by Sportspersons and the General Public of North Carolina.(2010-10-15) Dalrymple, Carolyn; Markus Peterson, Committee Chair; David Cobb, Committee Member; Erin Sills, Committee Member; Howard Bondell, Committee Member
- Holiday Effects on Retail Consumptions in the U.S. Economy.(2010-05-04) Gao, Junlin; Michael Walden, Committee Chair; Lee Craig, Committee Member; Howard Bondell, Committee Member
- Multivariate Robust Estimation of DCC-GARCH Volatility Model.(2010-05-07) LaBarr, Aric; Peter Bloomfield, Committee Chair; Howard Bondell, Committee Member; Denis Pelletier, Committee Member; David Dickey, Committee Member
- Random Effect Selection in Linear Mixed Models.(2010-08-16) Ahn, Mihye; Hao Zhang, Committee Chair; Wenbin Lu, Committee Chair; Daowen Zhang, Committee Member; Howard Bondell, Committee Member
- Relationship Lending and Lines of Credit for Small Business(2010-04-12) Gong, Jie; Douglas K. Pearce, Committee Chair; Karlyn Mitchell, Committee Member; Howard Bondell, Committee MemberThis thesis examines the influences of bank-borrower relationships on the terms for bank lines of credit for small business. I use the Surveys of Small Business Finances data to estimate two models: an OLS Regression explaining the premium over the prime rate and a Logistic Regression for the probability of collateral requirements. I focus on those firms with lines of credit with floating rates from commercial banks and use contract, financial, governance, industry and relationship characteristics as explanatory variables. Dun and Bradstreet (D&B) credit scores, minority status and gender are also added to previous models reported in the literature. My results are: (1) Small firms with longer market experiences will pay lower premium rates over the prime rate and firms with higher risk D&B credit scores will pay higher premiums. These results are both statistically and economically significant. However, the length of bank-borrower relationships does not have a statistically significant effect on the loan rate. Although lines of credit may provide more ‘soft-information’ on borrowers during bank-borrower relationships, banks still put more weight on credit scores and the firms’ age. (2) There is no statistically significant relationship between Relationship Characteristics and the probability of collateral requirements. Banks pay more attention to Financial Characteristics and type of ownership. D&B credit scoring system plays a more important role than bank-borrower relationship status. (3) Minority status and gender do not have impacts on loan rates or the probability of pledging collateral.
- Variable Selection Methods with Applications to Shape Restricted Regression(2008-08-04) Curtis, Steven McKay; Helen Zhang, Committee Member; Howard Bondell, Committee Member; Sujit K. Ghosh, Committee Co-Chair; Subhashis Ghosal, Committee Co-Chair
