AuthorErgun, Ahmet T.
AdvisorKer, Alan P.
MetadataShow full item record
PublisherThe University of Arizona.
RightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
AbstractThis dissertation focuses on econometric methodology and its applications in insurance and the stock market. The second chapter proposes a new semiparametric estimator for binary-choice single-index models. The estimator makes use of a "parametric start" idea from the statistics literature and applies it to econometric model estimation. Even though the chapter only focuses on binary-choice models, it is expected that the introduction of this idea to the econometrics literature is going to contribute to semiparametric estimation of econometric models in general, especially when one has (only) a rough initial guess about the shape of the unknown function. Consistency of the estimator is shown and the simulation results indicate that even though the parametric start is not correct in any of the simulation designs, the estimator's performance is very promising in the estimation of coefficients and probabilities. The third chapter successfully applies this proposed estimator along with competing parametric and semiparametric estimators and is expected to expand our understanding of private insurance company involvement in the U.S. crop insurance program. This chapter stands almost alone in the literature as an overwhelming majority of other studies examine the involvement of producers in the program. Although preliminary, the results of this chapter show that the insurance company involvement in this program may be too costly to justify and that the program may not be as efficient in terms of premium rates and rating practices of the federal government. The fourth chapter examines market volatility taking into account the New York Stock Exchange trading collar. The trading collar restricts certain forms of trade in component stocks of the S&P500 stock price index when there is "excess" volatility in the market. This important feature of the market has been ignored in the large volatility modeling literature and it is expected that this chapter contributes to this literature by showing that after some data manipulation it is straightforward to incorporate this feature into standard econometric models. Another contribution of this chapter is the successful use of a polynomial specification to capture the well documented U-shaped pattern of intraday market volatility instead of a computationally more difficult two-step procedure.
Degree ProgramGraduate College