In this thesis, we present a decision-theoretic framework for building decision support systems that incrementally elicit preferences of individual users over multiattribute outcomes and then provide recommendations based on the acquired preference information. By combining decision-theoretically sound modeling with effective computational techniques and certain user-centric considerations, we demonstrate the feasibility and potential of practical autonomous preference elicitation and recommendation systems. More concretely, we focus on decision scenarios in which a user can obtain any outcome from a finite set of available outcomes. The outcome is space is multiattribute; each outcome can be viewed as an instantiation of a set of attributes with finite domains. The user has preferences over outcomes that can be represented by a utility function. We assume that user preferences are generalized additively independent (GAI), and, therefore, can be represented by a GAI utility function. GAI utilities provide a flexible representation framework for structured preferences over multiattribute outcomes; they are less restrictive and, therefore, more widely applicable than additive utilities. In many decision scenarios with large and complex decision spaces (such as making travel plans or choosing an apartment to rent from thousands of available options), selecting the optimal decision can require a lot of time and effort on the part of the user. Since obtaining the user's complete utility function is generally infeasible, the decision support system has to support recommendation with partial preference information. We provide solutions for effective elicitation of GAI utilities in situations where a probabilistic prior about the user's utility function is available, and in situations where the system's uncertainty about user utilities is represented by maintaining a set of feasible user utilities. In the first case, we use Bayesian criteria for decision making and query selection. In the second case, recommendations (and query strategies) are based on the robust minimax regret criterion which recommends the outcome with the smallest maximum regret (with respect to all adversarial instantiations of feasible utility functions). Our proposed framework is implemented in the UTPref recommendation system that searches multiattribute product databases using the minimax regret criterion. UTPref is tested with a study involving 40 users interacting with the system. The study measures the effectiveness of regret-based elicitation, evaluates user comprehension and acceptance of minimax regret, and assesses the relative difficulty of different query types.

Thesis (Ph. D.)--Ohio State University, 2004. / Title from first page of PDF file. Document formatted into pages; contains xii, 138 p.; also includes graphics. Includes abstract and vita. Advisor: Greg M. Allenby, Dept. of Business Admnistration. Includes bibliographical references (p. 134-138).

Thesis (Ph. D.)--University of Missouri-Columbia, 2002. / Typescript. Vita. Includes bibliographical references (leaves 192-195). Also available on the Internet.

Thesis (Ph. D.)--University of Missouri-Columbia, 2002. / Typescript. Vita. Includes bibliographical references (leaves 192-195). Also available on the Internet.

This dissertation presents three essays. The first essay, coauthored with Tomasz Strzalecki, is a classical exercise in axiomatic decision theory. We propose a simple and novel axiomatization of quasi-hyperbolic discounting, a tractable model of present bias preferences that has found many applications in economics. Our axiomatization imposes consistency restrictions directly on the intertemporal tradeoffs faced by the decision maker, without relying on auxiliary calibration devices such as lotteries. Such axiomatization is useful for experimental work since it renders the short-run and long-run discount factor elicitation independent of assumptions on the decision maker's utility function. The second essay, coauthored with Carolin Pflueger, belongs to the field of econometric theory. We develop a test for weak identification in the context of linear instrumental variables regression. The central feature of our test is its robustness to heteroskedasticity, autocorrelation, and clustering. We define identification to be weak when the Two-Stage Least Squares (TSLS) or the Limited Information Maximum Likelihood (LIML) Nagar bias is large relative to a benchmark. To test the null hypothesis of weak identification we propose a scaled non-robust first stage F statistic: the effective F. The test rejects for large values of the effective F. The critical values depend on an estimate of the covariance matrix of the OLS reduced form regression coefficients and on the covariance matrix of the reduced form errors. The third essay—the main chapter of this dissertation—belongs to the intersection of econometric theory and statistical decision theory. I present a new class of tests for hypothesis testing problems with a special feature: a boundary-sufficient statistic. The new tests minimize a weighted sum of the average rates of Type I and Type II error (average risk), while controlling the conditional rejection probability on the boundary of the null hypothesis; in this sense they are efficient conditionally similar on the boundary (ecs). The ecs tests emerge from an axiomatic approach: they essentially characterize admissibility—an important finite-sample optimality property—and similarity on the boundary in the class of all tests, provided the boundary-sufficient statistic is boundedly complete. / Economics

No description available.

Due to the nature of survey design, the estimation of parameters associated with small areas is extremely problematic. In this study, techniques for the estimation of small area proportions are proposed and implemented. More specifically, empirical Bayes estimation methodologies, where random effects which reflect the complex structure of a multi-stage sample design are incorporated into logistic regression models, are derived and studied. / The proposed techniques are applied to data from the 1950 United States Census to predict local labor force participation rates of females. Results are compared with those obtained using unbiased and synthetic estimation approaches. / Using the proposed methodologies, a sensitivity analysis concerning the prior distribution assumption, conducted with a view toward outlier detection, is performed. The use of bootstrap techniques to correct measures of uncertainty is also studied.

Project (M.S.)--Brigham Young University. Dept. of Statistics, 2005. / Includes bibliographical references (p. 108-109).

Thesis (M.S.) -- Worcester Polytechnic Institute. / Keywords: industry and occupation codes; Census classification; Bayesian. Includes bibliographical references (p.).

Thesis (Ph.D.)--Baylor University, 2006. / Includes bibliographical references (p. 123-128).