Doctoral Dissertations

Permanent URI for this collection

Browse

Recent Submissions

Now showing 1 - 5 of 814
  • Item
    Essays on the Airbnb Market: Superhosts and Local Policy Determinants and Effects
    (Middle Tennessee State University, 2021) Smith, Joshua Philip ; Roach, Michael ; Rennhoff, Adam ; Sprick Schuster, Steven
    This dissertation is composed of three distinct empirical analyses, separated by chapter. Chapter I examines the impact that salient information has on host and guest decisions in the sharing economy using data from Airbnb from December 2018 to April 2019. Airbnb’s Superhost badge offers a shortcut to consumers searching for high-quality sellers. By estimating the impact of acquiring the badge separately from the conditions that Airbnb uses to merit Superhosts, I isolate the effect of salience of the information the badge carries via OLS and fixed effect panel models. The result is a negligible increase in price by sellers, but a growing effect on the number of reviews, 0.10 and 0.27, one and two months after earning the badge, respectively. I estimate the effect on revenues is between 10% to 17% increase from increased bookings by guests. Chapter II asks why local governments pass restrictions on short-term rentals, such as Airbnb. I construct a novel classification of these laws passed by cities. I use panel binary probits and ordered models to predict the marginal effects of local economic conditions on short-term rental restrictions using data from 2012 to 2019 in nineteen U.S. cities. I find that a one standard deviation decline in housing affordability leads to a 20.57 percentage point increase in the likelihood that a city council restricts in a specific approach that only personal residences may be operated as a short-term rental. Alternatively, a one standard deviation increase in affordability predicts a 23.78 percentage point increase in the likelihood that no restrictive policy is passed. Chapter III identifies the effect of policies aimed at reducing short-term rental supply. I use fixed effects panel models with five years of data from Airbnb to show the casual response of professional and nonprofessional suppliers in the United States from changes enacted through city law. I find these restrictions do little to reduce professional supply but significantly affect nonprofessional hosts, reducing availability by as much as -15.7%. A 1 percent increase in permit fees leads to a -0.7 percentage point decrease in professional's supply. Yet for nonprofessionals, the same 1 percent increase leads to a percentage point increase of 1 to 1.2 percentage points in supply. My paper expands the limited and conflicting empirical research of local policy aimed at short-term rentals by offering robust methods to disentangle the heterogeneous effects by host type.
  • Item
    APPLICATIONS OF MODERN NLP TECHNIQUES FOR PREDICTIVE MODELING IN ACTUARIAL SCIENCE
    (Middle Tennessee State University, 2021) Xu, Shuzhe ; Hong, Don ; Barbosa, Salvador E. ; Manathunga, Vajira ; Sinkala, Zachariah ; Wu, Qiang
    In this dissertation, the research focuses on Natural Language Processing (NLP) applications in actuarial science. NLP techniques, as powerful text analytic tools, can automatically help actuaries to exploit the information in textual data. Recently, many NLP techniques have been applied in different research fields, but only a few NLP applications can be found in actuarial science. This dissertation researches NLP techniques in actuarial science and proposes some NLP solutions for actuarial applications. This dissertation consists of five chapters. The first chapter is an introduction of NLP and some opportunities for its use in actuarial science. The possibilities of traditional actuarial applications incorporating NLP are also discussed. A few NLP applications proposed by actuaries are also introduced as references. The second chapter is the literature review of relevant NLP techniques. Some basic technologies are introduced such as word embeddings and tokenizations. Also, advanced NLP tools such as Bidirectional Encoder Representation for Transformers (BERT) and related techniques are discussed. The third chapter is an NLP application based on extended truck warranty data. This chapter develops a BERT-based aggregate loss model with a rescaled 10-value scale severity to predict future losses based on the frequency distribution of claim counts with contracts and severity distribution of claim records. The NLP tool helps to extract information from the textual description in the data, and the extracted values are exploited to predict loss severity. The fourth chapter is another NPL application for basic truck warranty data. A data-based portfolio allocation model is proposed to predict losses using the modern portfolio theory (MPT) developed by Nobel Laureate Harry Markowitz in 1952. In this chapter, BERT is applied to improve the accuracy of multi-class classification in the BERT enhanced data-based portfolio allocation model. Also, a technique similar to the one used in chapter 3 is applied to derive a BERTbased severity model for multi-class aggregate loss prediction through a different approach with the BERT enhanced data-based portfolio allocation model. The last chapter summarizes the described applications. The applications of modern NLP techniques for predictive analytics are practical and promising. However, applications to actuarial science are almost nonexistent. This dissertation demonstrates the possibilities of NLP applications to improve predictive modeling in actuarial science. The NLP techniques can help to gather information from textual descriptions discarded by traditional models. The possible improvements that can be made in future research are also described in this chapter.
  • Item
    APPLICATIONS OF MODERN NLP TECHNIQUES FOR PREDICTIVE MODELING IN ACTUARIAL SCIENCE
    (Middle Tennessee State University, 2021) Xu, Shuzhe ; Hong, Don ; Barbosa, Salvador E. ; Manathunga, Vajira ; Sinkala, Zachariah ; Wu, Qiang
    In this dissertation, the research focuses on Natural Language Processing (NLP) applications in actuarial science. NLP techniques, as powerful text analytic tools, can automatically help actuaries to exploit the information in textual data. Recently, many NLP techniques have been applied in different research fields, but only a few NLP applications can be found in actuarial science. This dissertation researches NLP techniques in actuarial science and proposes some NLP solutions for actuarial applications. This dissertation consists of five chapters. The first chapter is an introduction of NLP and some opportunities for its use in actuarial science. The possibilities of traditional actuarial applications incorporating NLP are also discussed. A few NLP applications proposed by actuaries are also introduced as references. The second chapter is the literature review of relevant NLP techniques. Some basic technologies are introduced such as word embeddings and tokenizations. Also, advanced NLP tools such as Bidirectional Encoder Representation for Transformers (BERT) and related techniques are discussed. The third chapter is an NLP application based on extended truck warranty data. This chapter develops a BERT-based aggregate loss model with a rescaled 10-value scale severity to predict future losses based on the frequency distribution of claim counts with contracts and severity distribution of claim records. The NLP tool helps to extract information from the textual description in the data, and the extracted values are exploited to predict loss severity. The fourth chapter is another NPL application for basic truck warranty data. A data-based portfolio allocation model is proposed to predict losses using the modern portfolio theory (MPT) developed by Nobel Laureate Harry Markowitz in 1952. In this chapter, BERT is applied to improve the accuracy of multi-class classification in the BERT enhanced data-based portfolio allocation model. Also, a technique similar to the one used in chapter 3 is applied to derive a BERTbased severity model for multi-class aggregate loss prediction through a different approach with the BERT enhanced data-based portfolio allocation model. The last chapter summarizes the described applications. The applications of modern NLP techniques for predictive analytics are practical and promising. However, applications to actuarial science are almost nonexistent. This dissertation demonstrates the possibilities of NLP applications to improve predictive modeling in actuarial science. The NLP techniques can help to gather information from textual descriptions discarded by traditional models. The possible improvements that can be made in future research are also described in this chapter.
  • Item
    APPLICATIONS OF MODERN NLP TECHNIQUES FOR PREDICTIVE MODELING IN ACTUARIAL SCIENCE
    (Middle Tennessee State University, 2021) Xu, Shuzhe ; Hong, Don ; Barbosa, Salvador E. ; Manathunga, Vajira ; Sinkala, Zachariah ; Wu, Qiang
    In this dissertation, the research focuses on Natural Language Processing (NLP) applications in actuarial science. NLP techniques, as powerful text analytic tools, can automatically help actuaries to exploit the information in textual data. Recently, many NLP techniques have been applied in different research fields, but only a few NLP applications can be found in actuarial science. This dissertation researches NLP techniques in actuarial science and proposes some NLP solutions for actuarial applications. This dissertation consists of five chapters. The first chapter is an introduction of NLP and some opportunities for its use in actuarial science. The possibilities of traditional actuarial applications incorporating NLP are also discussed. A few NLP applications proposed by actuaries are also introduced as references. The second chapter is the literature review of relevant NLP techniques. Some basic technologies are introduced such as word embeddings and tokenizations. Also, advanced NLP tools such as Bidirectional Encoder Representation for Transformers (BERT) and related techniques are discussed. The third chapter is an NLP application based on extended truck warranty data. This chapter develops a BERT-based aggregate loss model with a rescaled 10-value scale severity to predict future losses based on the frequency distribution of claim counts with contracts and severity distribution of claim records. The NLP tool helps to extract information from the textual description in the data, and the extracted values are exploited to predict loss severity. The fourth chapter is another NPL application for basic truck warranty data. A data-based portfolio allocation model is proposed to predict losses using the modern portfolio theory (MPT) developed by Nobel Laureate Harry Markowitz in 1952. In this chapter, BERT is applied to improve the accuracy of multi-class classification in the BERT enhanced data-based portfolio allocation model. Also, a technique similar to the one used in chapter 3 is applied to derive a BERTbased severity model for multi-class aggregate loss prediction through a different approach with the BERT enhanced data-based portfolio allocation model. The last chapter summarizes the described applications. The applications of modern NLP techniques for predictive analytics are practical and promising. However, applications to actuarial science are almost nonexistent. This dissertation demonstrates the possibilities of NLP applications to improve predictive modeling in actuarial science. The NLP techniques can help to gather information from textual descriptions discarded by traditional models. The possible improvements that can be made in future research are also described in this chapter.
  • Item
    ANTIFUNGAL ASSESSMENTS, CHEMOGENOMIC AND TRANSCRIPTOMIC ANALYSES, AND STRUCTURE-ACTIVITY RELATIONSHIPS OF BIOACTIVE AURONE COMPOUNDS
    (Middle Tennessee State University, 2021) Alqahtani, Fatmah Mohammed ; Farone, Mary B ; Handy, Scott T ; Farone, Anthony L ; Seipelt-Thiemann, Rebecca L ; Kline, Paul C
    Candida spp. as commensal colonizers of mucocutaneous surfaces in humans are the major fungal cause of bloodstream infections, resulting in 50,000 deaths every year. Because of the paucity of fundamental antifungal treatments along with the recurrent emergence of resistant strains, the urgency for new antifungal agents with selective and novel targets necessitates the efforts of laboratory research. Complementary to such attempts for identification of novel anti-Candida agents, two synthetically aurones, SH1009 and SH9051, have shown significant inhibitory activities against Candida spp. The main goal of this research was to assess antifungal activity and mammalian cell cytotoxicity and comprehensively characterize the modes of action for aurones SH1009 and SH9051. Aurone SH1009 exhibited significant antifungal activity against Candida spp., including resistant isolates, with fungistatic pharmacodynamic properties. SH1009 treatment of a pooled genome-wide set of Saccharomyces cerevisiae deletion mutants demonstrated a set of sensitive and resistant growth responses in mutants encoding cell cycle-dependent organization of actin cytoskeleton. Accordingly, phenotypic studies revealed cell cycle arrest at the G1 phase in SH1009-treated Candida albicans along with abnormally interrupted actin dynamics and enlarged, unbudded cells, validating the chemical genetic interaction and suggesting a novel mode of action for aurone SH1009 as an antifungal. In vitro cytotoxicity assays using human cell lines showed selective toxicity of SH1009 toward fungal cells and less selectivity in SH9051-treated cells. In an attempt to increase antifungal activity by combining both aurones, antifungal synergy was detected as an indifferent interaction, suggesting different modes of action for the aurones. To interpret these differences, transcriptome changes in SH1009- and SH9051-treated C. albicans were analyzed, revealing common and unique pathways enriched consistently based on the chemical structure of each aurone. A reverse genetic approach coupled with structure-activity relationship analyses demonstrated that trehalose was uniquely responsible for SH1009 resistance. SH9051 uniquely stimulated sulfur amino acid catabolism, and the core chemical structure of both aurones commonly promoted intracellular oxidative stress. The results of these studies determined a selectively novel molecular target, for aurone SH1009 as a promising antifungal and revealed cellular effects for different functional groups of the aurone, paving the way for future development and investigation.