Rejection sampling and importance sampling and their roles in MCMC. Limited to Master of Applied Statistics students. Graphics and real examples used to illustrate techniques. Topics include Stein paradox, nonparametric Bayes, and statistical learning. Designed for social sciences graduate students and advanced undergraduate students seeking training in data issues and methods employed in social sciences. Simulated annealing. Requisite: course 200B. Jun-88, Costly Adjustment Under Rational Expectations: A Generalization, UCLA, University of New Mexico   Introduction to state-of-art statistical methods that rely on historical data collected in past to forecast future outcomes. Principal components, canonical correlation, discriminant analysis.

Preparation: basic statistics, linear algebra (matrix analysis), computer vision. May-91, UCLA, University of Aalborg (Denmark)   Simulation, renewal theory, martingale, and selected topics from queuing, reliability, speech recognition, computational biology, mathematical finance, epidemiology. Study of four commonly employed solutions--SPSS (Statistical Package for Social Sciences), Stata, SAS (Statistical Analysis System), and R--for data analytic and statistical issues in health sciences, engineering, economics, and government. Recommended: course 102A. Probability distributions, random variables, vectors, and expectation. Lecture, three hours. Feb-05, Philippe Jehiel, Morita Meyer-ter-Vehn, Benny Moldovanu, William R. Zame, UCL London and PSE Paris,University of Bonn,University of California, Los Angeles   Requisites: courses 10, 20, and 101A, or equivalent level of discipline. Concurrently scheduled with course C225. Letter grading. Causal modeling: theory testing via analysis of moment structures.

Overview of fundamental concepts of data analysis and statistical inference and how these are applied in wide variety of settings. Use of Statistics Online Computational Resource (SOCR). Seminar, three hours. Seminar, two hours.

P/NP or letter grading. Students learn how to lead, manage, negotiate, and participate in teams of data scientists. Lecture, four hours. Designed for upper-division and graduate students in social or life sciences and those who plan to major in Statistics. Limited to junior/senior USIE facilitators. Generalized linear model and maximum likelihood methods as essential tools all statistics students should understand.

To further knowledge by applying what students have learned in class to an actual service work setting under guidance of faculty mentor. Requisite: course 202B.