Thesis: Ph. D., Massachusetts Institute of Technology, Department of Economics, 2018. / Cataloged from PDF version of thesis. / Includes bibliographical references (pages 153-154). / This thesis consists of three essays on search for reviews. In Chapter 1, I study how homogeneous consumers behave when they are considering to buy a new experience good, such as a pair of shoes, from a monopolist with an observable price. They have costly access to consumer reviews which are perfectly or imperfectly informative of match quality. Knowing how consumers behave, the monopolist sets an optimal price inducing certain behavior; sometimes the firm will find it optimal to set a high price to induce the consumers to search for earlier reviews, sometimes a low price to induce them to purchase the product. Consumer, producer and social surplus are non-monotone in search cost, and this result extends to settings with heterogeneous consumers. In Chapter 2, we extend the model of Chapter 1 to allow for competition between ex-ante identical firms. Consumers are homogeneous and all prices observed at no cost. They can search for earlier reviews which perfectly reveal match quality. Consumers can keep searching as long as they have not found a match or exhausted all of their options. We learn that high search costs lead to relatively high prices but when costs decrease, surplus is first transferred from firms to consumers but further reductions may decrease consumer surplus. Additional effects are noted as reviews get even cheaper to access, and consumer surplus, profits, and social surplus are all non-monotone in the cost of reading reviews. In Chapter 3, we analyze selection into consumption in the case of movies. People leaving reviews are assumed to be a representative sample of consumers, so that reviews perfectly reveal experiences. New consumers observe these reviews but do not know if their preferences are aligned with those of the reviewers. Examining two datasets with movie reviews and box office revenue, both in a cross-section of movies and within movies over time, we learn that selection decreases in the expected quality of a movie, the precision of the prior quality, and consumer homogeneity. Selection may increase or decrease over time and it tends to increase in the number of movies. Thesis Supervisor: Glenn Ellison Title: Gregory K. Palm Professor of Economics Thesis Supervisor: Michael Whinston Title: Sloan Fellows Professor of Management 3 / by Kosti Takala. / Ph. D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Economics, 2013. / Cataloged from PDF version of thesis. / Includes bibliographical references (p. 201-208). / Monotonicity is a key qualitative prediction of a wide array of economic models derived via robust comparative statics. It is therefore important to design effective and practical econometric methods for testing this prediction in empirical analysis. Chapter 1 develops a general nonparametric framework for testing monotonicity of a regression function. Using this framework, a broad class of new tests is introduced, which gives an empirical researcher a lot of flexibility to incorporate ex ante information she might have. Chapter 1 also develops new methods for simulating critical values, which are based on the combination of a bootstrap procedure and new selection algorithms. These methods yield tests that have correct asymptotic size and are asymptotically nonconservative. It is also shown how to obtain an adaptive rate optimal test that has the best attainable rate of uniform consistency against models whose regression function has Lipschitz-continuous first-order derivatives and that automatically adapts to the unknown smoothness of the regression function. Simulations show that the power of the new tests in many cases significantly exceeds that of some prior tests, e.g. that of Ghosal, Sen, and Van der Vaart (2000). An application of the developed procedures to the dataset of Ellison and Ellison (2011) shows that there is some evidence of strategic entry deterrence in pharmaceutical industry where incumbents may use strategic investment to prevent generic entries when their patents expire. Many economic models yield conditional moment inequalities that can be used for inference on parameters of these models. In chapter 2, I construct a new test of conditional moment inequalities based on studentized kernel estimates of moment functions. The test automatically adapts to the unknown smoothness of the moment functions, has uniformly correct asymptotic size, and is rate optimal against certain classes of alternatives. Some existing tests have nontrivial power against n-1/2 -local alternatives of a certain type whereas my method only allows for nontrivial testing against (n/ log n)-1/2-local alternatives of this type. There exist, however, large classes of sequences of well-behaved alternatives against which the test developed in this paper is consistent and those tests are not. In chapter 3 (coauthored with Victor Chernozhukov and Kengo Kato), we derive a central limit theorem for the maximum of a sum of high dimensional random vectors. Specifically, we establish conditions under which the distribution of the maximum is approximated by that of the maximum of a sum of the Gaussian random vectors with the same covariance matrices as the original vectors. The key innovation of this result is that it applies even when the dimension of random vectors (p) is large compared to the sample size (n); in fact, p can be much larger than n. We also show that the distribution of the maximum of a sum of the random vectors with unknown covariance matrices can be consistently estimated by the distribution of the maximum of a sum of the conditional Gaussian random vectors obtained by multiplying the original vectors with i.i.d. Gaussian multipliers. This is the multiplier bootstrap procedure. Here too, p can be large or even much larger than n. These distributional approximations, either Gaussian or conditional Gaussian, yield a high-quality approximation to the distribution of the original maximum, often with approximation error decreasing polynomially in the sample size, and hence are of interest in many applications. We demonstrate how our central limit theorem and the multiplier bootstrap can be used for high dimensional estimation, multiple hypothesis testing, and adaptive specification testing. All these results contain non-asymptotic bounds on approximation errors. / by Denis Chetverikov. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Economics, 1996. / Includes bibliographical references (leaves 96-100). / by Patrizia Canziani. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Economics, 1986. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND DEWEY. / Includes bibliographies. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Economics, 1991. / Includes bibliographical references (p. 722-745). / by Stephen Herzenberg. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Economics, February 2007. / Includes bibliographical references. / This dissertation is composed of three chapters, the first demonstrates that natural gas violates many of the simplifying assumptions frequently used in modeling its behavior. Careful analysis of futures contracts written on gas suggests that gas prices are seasonal while returns are non-Gaussian and evidence stochastic volatility. In addition, examination of options prices indicates the intermittent presence of jumps. We find that models which disregard these properties struggle to recover options prices with any precision. Thus, we propose an alternative nonparametric approach to gas options pricing that captures these salient features while also shedding light on the nature of risk aversion embedded in gas markets. The second chapter presents new estimates and approaches to estimating the home bias puzzle. It uses micro-level data to calculate households' foreign equity exposure as a function of wealth. We find simple estimates have significant errors-in-variables problems and we construct an estimator using grouping to account for this issue. Our estimates still imply low aggregate investment in foreign equity. Finally, we disaggregate the investment decision by incorporating two step decisions that allow households to forgo participating in the market. / (cont.) As a result of the decoupling, we find foreign equity levels closer to that of standard portfolio theories. The final chapter considers principal-agent models in which the principal cannot measure the output nor the effort level of agents. To model this situation, we use utility models that include identity, justified partly by empirical results from peer-effects, and apply these extended utility functions. In the single agent case, introducing identity amounts to modifying the utility function and does not lead to dramatic results. In the multiple agent case, we find that the addition of identity can lead to more efficient outcomes than cases where identity is ignored. The addition of identity, however, can also lead to counter-intuitive results due to the interactions among agents and may produce second-best outcomes that are worse than the case without identity. Finally the addition of identity can help explain some empirical results that may be difficult to explain with standard models. / by Alan Michael Grant. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Economics, 2007. / Includes bibliographical references. / This thesis consists of three essays on the economics of education.The first chapter estimates the effects of participating in the National School Lunch Program in the middle of the 20th century on educational attainment and adult health. My instrumental variables strategy exploits a change of the formula used by the federal government to allocate funding to the states that was phased in beginning in 1963. Identification is achieved by the fact that different birth cohorts were exposed to different degrees to the original formula and the new formula, along with the fact that the change of the formula affected states differentially by per capita income. Participation in the program as a child appears to have few long-run effects on health, but the effects on educational attainment are sizable. The second chapter studies the issue of racial diversity in higher education. I estimate the effects of college racial diversity on post-college earnings, civic behavior, and satisfaction with the college attended. I use the Beginning Postsecondary Students survey, which allows me to control for exposure to racial diversity prior to college. Moreover, I use two techniques from Altonji, Elder, and Taber (2005) to address the issue of selection on unobservables. Single-equation estimates suggest a positive effect of diversity on voting behavior and on satisfaction with the college attended, but I do not find an effect on other outcomes. Moreover, the estimates are very sensitive to the assumptions made about selection on unobservables.The third chapter studies university affirmative action bans. I use information on the timing of bans along with data from the Current Population Survey (CPS) and the American Community Survey (ACS) to estimate the effects of such bans on college enrollment and educational attainment. / (cont.) I use a triple difference strategy that uses whites as a comparison group for underrepresented minorities and that exploits variation in the bans over states and across time. I find no adverse impact of bans on overall minority college attendance rates and educational attainment relative to whites, and I find no effect of the bans on minority enrollment in public colleges or four-year colleges. / by Peter Hinrichs. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Economics, 1990. / Includes bibliographical references. / by Leonardo Felli. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Economics, 1993. / Includes bibliographical references. / by David Morris Frankel. / Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Economics, 1987. / Bibliography: leaves 208-224. / by Daniel Martin Gorodetsky Raff. / Ph.D.