Course Announcements

Course Announcements Archive
Last revised: 8/08

SPRING 2008

College Courses

STATISTICS 20000. Elementary Statistics.
Sec 01: Dan Wang, MWF 9:30-10:20 AM, Eckhart 133.
Sec 02: Wenlong Wang, MWF 12:30-1:20 PM, Eckhart 133.
PQ: Math 10500 or equivalent.
Required reading: Statistics, 4th edition, by Freedman, Pisani, Purves 2007, Norton.
ISBN-10: 0393929728, ISBN-13: 978-0393929720.

This course meets one of the general education requirements in the mathematical sciences. NOTE: STAT 20000 may not be used in the statistics major. It is recommended for students who do not plan to take advanced statistics courses. This course introduces statistical concepts and methods for the collection, presentation, analysis, and interpretation of data. Elements of sampling, simple techniques for analysis of means, proportions, and linear association are used to illustrate both effective and fallacious uses of statistics.

STATISTICS 22000. Statistical Methods and Their Applications.
Sec 01: Zuoheng Wang, MWF 10:30-11:20 AM, Harper Memorial 140.
Sec 02: Shali Wu, MWF 1:30-2:20 PM, Eckhart 133.
PQ: 2 QTRS Calculus.
Required reading: Introduction to the Practice of Statistics, 5th edition by Moore and McCabe
2006, W. H. Freeman. ISBN-10: 0716764008, ISBN-13: 978-0716764007

This course introduces statistical techniques and methods of data analysis, including the use of computers. Examples are drawn from the biological, physical, and social sciences. Students are required to apply the techniques discussed to data drawn from actual research. Topics include data description, graphical techniques, exploratory data analyses, random variation and sampling, one- and two-sample problems, the analysis of variance, linear regression, and analysis of discrete data.

STATISTICS 22200 Linear Models and Experimental
Sec 01: Linda Collins, TTh, 9:00-10:20 AM, Eckhart 133.
PQ: STAT 22000 or consent of instructor.
Required Reading: Oehlert, G. W. (2000) A First Course in Design and Analysis of Experiments. W. H. Freeman. ISBN-10: 0-7167-3510-5 ISBN-13: 978-0-7167-3510-6.

This course covers principles and techniques for the analysis of experimental data and the planning of the statistical aspects of experiments. Topics include linear models, analysis of variance, randomization, blocking, factorial designs, confounding, and incorporation of covariate information.

In addition to regular homework assignments and exams, students will be required to complete a project involving the design and analysis of an experiment of their own.

STATISTICS 23400. Statistical Models and Methods.
Sec 01: Pending, MWF, 11:30-12:20 PM, Eckhart 133.
Sec 02: Michael Finegold, MWF, 2:30-3:20 PM, Eckhart 133.
PQ: Mathematics 13300, 15300 or 16300.
Required Reading: Investigating Statistical Concepts, Applications, and Methods by Chance and Rossman 2006, Duxbury (Thomson Brooks/Cole). ISBN-10: 0495050644, ISBN-13: 978-D495050643.

A Brief Course in Mathematical Statistics by Tanis and Hogg 2007, Prentice Hall, ISBN-10: 0131751395, ISBN-13: 978-0131751392.

This course presents basic ideas of probability theory and statistics, and is recommended for students throughout the natural and social sciences who want a broad background in statistical methodology and exposure to probability models and the statistical concepts underlying the methodology. Probability is developed for the purpose of modeling outcomes of random phenomena. Random variables and their expectations are studied; including means and variances of linear combinations, and an introduction to conditional expectation. Binomial, hypergeometric, Poisson, exponential, normal and other standard probability distributions are considered. Some probability models are studied mathematically and others via simulation on a computer. Sampling distributions and related statistical methods are explored mathematically, studied via simulation and illustrated on data. Statistical methods for describing data and making inferences based on samples from populations are presented. Methods include, but are not limited to, inference for proportions and means for one- and two-sample problems, correlation and simple linear regression. Graphical and numerical data description are used for exploration, communication of results, and comparing mathematical consequences of probability models and data. Mathematics is employed to the level of univariate calculus and is less demanding than that required by STAT 24400.

Univariate calculus and computer simulation are used throughout the course to investigate statistical concepts and their mathematical underpinnings. One full year of univariate calculus is a prerequisite for the course (Math 13300, 15300, or 16300). Familiarity with at least limits, derivatives and integrals of polynomial and exponential functions, change of variable (substitution) in definite integrals, max-min problems, use of summation notation, and sequences and series as well as a willingness to explore ideas mathematically are key to your success in this course. See http://statistics.uchicago.edu/~stat234 for more detailed information.

Stat 23400 takes the mathematical prerequisite quite seriously. Enrolling concurrently in either Math 13300, 15300, or 16300 while taking Stat 23400 is very strongly discouraged. Further, students who do not feel strong mathematically, may want to wait until completing their entire mathematical requirement (e.g., Math 19500-19600 for Economics majors) before enrolling in Stat 23400. Economics majors are strongly encouraged to delay taking Stat 23400 until the quarter just before enrolling in their required econometrics course (Econ 21000), for which Stat 23400 is a prerequisite. Thus, delaying Stat 23400 until at least late in the second year or even early in the third year of the Economics degree program should not be considered unusual.

STATISTICS 24500. Statistical Theory and Methods 2.
Sec 01: Debashis Mondal, TTh, 1:30 PM, Eckhart 133.
PQ: STAT 23400 and STAT 23500 or STAT 24400 or consent of instructor.
Required reading: Mathematical Statistics and Data Analysis, 3rd edition, Rice, John A. 2007, Duxbury. ISBN-10: 0534399428, ISBN-13: 978-0534399429.

This is the second quarter of a two-quarter sequence. Enrollment in the second quarter alone is permitted, although not recommended. The first quarter covered the basics -- tools from probability and the elements of statistical theory. The second quarter will cover statistical methodology, including the data transformation, regression phenomena, t-tests, analysis of variance, linear regression, correlation, and some multivariate distribution theory. Some principles of data analysis will be introduced, and an attempt will be made to present ANOVA and regression in a unified framework. Much of the material is covered in chapters 10-12 and 14 of the text, but other viewpoints and derivations will be introduced as well. The computer will be used in the second quarter. Some mathematical maturity will be assumed, to the level of calculus. The computer will be used for data analysis and simulation.

STATISTICS 24600. Statistical Theory and Methods 3.
Sec 01: Yali Amit, TTh, 10:30-11:50 AM, Eckhart 133.
PQ: STAT 24400 and STAT 24500 or consent of instructor.
Required Reading: Pattern Recognition and Machine Learning, Bishop, Christopher M. 2006, Springer Verlag. ISBN-10: 0387310738, ISBN-13: 978-0387310732.

This course will introduce a variety of modern statistical methods. We will start with the description of
families of multivariate models - multivariate normal distribution, graphical models, log-linear models.
We will then discuss inference for such models including the EM algorithm, Bayesian inference and Monte-carlo methods. The main application of the models will be in classification problems - i.e the 'generative approach'. We will also introduce some modern discriminative approaches for classification.

STATISTICS 25100. Intro to Math Probability.
Sec 01: Michael J. Wichura, TTh, 12:00-1:20 PM, Eckhart 312.
PQ: MATH 20000 or MATH 20400 or consent of instructor.
Required Reading: A First Course in Probability, 7th edition, Ross, S. 2005, Pearson/Prentice Hall.
ISBN-10: 0131856626, ISBN-13: 978-0131856622.

The aim of the course is to provide an introduction to the concepts of probability. The course will cover the basic ideas used to describe aspects of randomness, such as events, random variables, independence, and conditional probability, with emphasis on the methods, calculation, and applications of probability. The topics treated are: combinatorics; probability models; rules of probability; conditional probability; independence; random variables; expectation and standard deviation; games of chance; common discrete distributions --- uniform, binomial, hypergeometric, geometric, negative binomial, and Poisson --- and their interrelationships; univariate and multivariate density and distribution functions, change of variable formulas for densities, common continuous distributions --- beta, Cauchy, chisquare, exponential, gamma, lognormal, normal, uniform --- and their interrelationships; moment generating functions, laws of large numbers and the central limit theorem.

For more information, please visit http://galton.uchicago.edu/~wichura/Stat251/courseinfo.html

STATISTICS 26100=STATISTICS 33600. Time Dependent Data.
Sec 01: MIchael Stein, TTh, 1:30-2:50 PM, Eckhart 202.
PQ: STAT 24400 or STAT 24500.
Required Reading: Time Series Analysis and Its Applications: With R Examples, 2nd edition
Shumway and Stoffer 2006, Springer. ISBN-10: 0387293175, ISBN-13: 978-0387293172.

This course considers the modeling and analysis of data that are ordered in time. The main focus will be on quantitative observations taken at evenly spaced intervals and will include both time-domain and spectral approaches. Time permitting, statistical approaches to other data types, such as categorical observations or point processes, will be considered.

STATISTICS 26700=CHSS 32900, HIPS 25600, STAT 36700.
Sec 01: Stephen Stigler, MWF, 9:30-10:20 AM, Eckhart 117.
PQ: A course in Statistics.
Required Reading: The History of Statistics: The Measurement of Uncertainty Before 1900.
Stigler, Stephen M. 1990, Belknap Press of Harvard University Press. ISBN-10: 067440341X,
ISBN-13: 978-0674403413. Other materials will be distributed in class or by web.

This course will cover topics in the history of statistics, from the eleventh century to the middle of the twentieth century. The emphasis will be upon the period 1650 to 1950, and upon the mathematical developments in the theory of probability and how they came to be used in the sciences, both to quantify uncertainty in observational data and as a conceptual framework for scientific theories. The course will include broad views of the development of the subject, and closer looks at specific people and investigations, including reanalyses of historical data. Topics will include: Early probability; Probability in seventeenth century medicine; Inverse probability in inference; Statistical methods in early geodesy; The introduction of least squares; Statistics in social science; Statistics in early biology and psychology; Simulation; Statistics and the evaluation of social programs; Maximum likelihood estimation; Statistics and nineteenth century forensic science; Statistics and medical science. Major figures who will be examined include Pascal, various Bernoullis, De Moivre, Laplace, Boscovich, Gauss, Quetelet, Galton, Edgeworth, Karl Pearson, Gosset, Fisher, Neyman.

Graduate Courses

STATISTICS 30200. Mathematical Statistics 2.
Sec 01: Wei Biao Wu, MW, 1:30-2:50 PM, Eckhart 117.
PQ: STAT 30100 or consent of instructor.
Required Reading: Mathematical Statistics, 2nd ed. 2003. Corr. 4th printing edition (October 5, 2007). Jun Shao, Springer. ISBN-10:0387953825, ISBN-13: 978-0387953823.

This course continues the development of mathematical statistics. Topics of importance include: statistical decision theory, admissability and the Neyman-Pearson lemma, formal hypothesis testing, comparison with conditional frequentist and Bayesian viewpoints, ancillarity, interval estimation, UMP tests and MLR, unbiased tests, score statistics, generalized likelihood ratio tests and asymptotics, minimax estimation, empirical Bayes and "shrinkage".

STATISTICS 31300. Introduction to Stochastic Processes 2.
Sec 01: Per A. Mykland, TTh, 10:30-11:50 AM, Eckhart 117.
PQ: STAT 31200 or consent of instructor.
Required reading:

STATISTICS 32200. Bayesian Data Analysis.
Sec 01: Matthew Stephens, MW, 1:30-2:50 PM, Eckhart 308.
PQ: Consent of instructor.
Reading: There is no required text, but "Bayesian Theory" by Bernardo and Smith is background reading.

This course is aimed at graduate students in statistics, and others with the necessary statistical background. We will assume familiarity with standard statistical distributions (e.g. Normal, Poisson, Binomial, Exponential), with the laws of probability, and concepts of statistical inference (maximum likelihood estimation, hypothesis testing, confidence intervals, etc), and basic familiarity with the R statistical package.

The course will cover foundations of Bayesian statistics, including axiomatic development, exchangeability, De Finetti's theorem, Jeffreys and improper priors, decision theory, Bayesian hypothesis testing and Bayes factors. Concepts will be illustrated mainly by instructive "toy" examples, where calculations can be done by hand. However, we will also study more complex, practical applications of Bayesian statistics. Although methods of computation will be discussed, the primary focus will be on concepts, and not on computation.

STATISTICS 33200=HSTD 43200. Causal Inference.
Sec 01: Tyler Vanderweele, TTh, 10:30-11:50 AM, BSLC arr.
PQ: STAT 22400-22600 or HSTD 32400-32700 or equivalent or consent of instructor.
Required reading:

STATISTICS 34700. Generalized Linear Models.
Sec 01: Peter McCullagh, TTh, 3:00-4:20 PM, Eckhart 133.
PQ: STAT 34300 or consent of instructor.
Required Reading: Generalized Linear Models, 2nd edition, McCullagh and Nelder 1990, Chapman & Hall/CRC. ISBN-10: 0412317605, ISBN-13: 978-0412317606.
Recommended Reading: Modern Applied Statistics with S, 4th edition, Venables and Ripley 2003, Springer. ISBN-10: 0387954570, ISBN-13: 978-0387954578.
Applied Statistics: Principles and Examples, Cox and Snell 19821 Chapman & Hall/CRC. ISBN-10: 0412165708, ISBN-13: 978-0412165702.

This is an applied course for students who are familiar with linear models at the level of Draper and Smith or Weisberg. The following topics will be covered:

Factors, variates, contrasts, interactions
Exponential-family models: variance function
Definition of a generalized linear model: link functions
Analysis of deviance
Specific examples of GLMs

logistic and probit regression
cumulative logistic models
log-linear models and contingency tables
inverse linear models

Quasi-likelihood and least squares; estimating functions
Over-dispersion
Partially linear models

STATISTICS 35500. Statistical Genetics.
Sec 01: Mary Sara McPeek, F, 1:30-4:10 PM, Eckhart 117.
PQ: Human Genetics 471 and Statistics 244 and 245. Students who do not meet the prerequisites may enroll on a P/NP basis with consent of the instructor.
Reading: There is no textbook.

This is an advanced course in statistical genetics. Prerequisites are Human Genetics 471 and Statistics 244 and 245. Students who do not meet the prerequisites may enroll on a P/NP basis with consent of the instructor. This is a discussion course and student presentations will be required. Topics vary and may include, but are not limited to, statistical problems in association mapping, linkage mapping, population genetics, microarray analysis, and genetic models for complex traits.

STATISTICS 36700=CHSS 32900, HIPS 25600, STAT 26700.
Sec 01: Stephen Stigler, MWF, 9:30-10:20 AM, Eckhart 117.
PQ: A course in Statistics.
Required Reading: Stephen M. Stigler, The History of Statistics: The Measurement of Uncertainty Before 1900. (Cambridge, Mass.: Harvard University Press, 1986.) (Available in paperback.) Other materials will be distributed in class or by web.

STATISTICS 38600. Topics in Stochastic Processes.
Sec 01: Steven P. Lalley, TTh, 3:00-4:20 PM, Eckhart 117.
PQ: Consent of instructor.
Recommended Reading: Stochastic Differential Equations by Oksendal (PG)
Brownian Motion and Stochastic Calculus by Karatzas and Shreve (R)
Foundations of Modern Probability by Kallenberg (X)

Topics:

Wiener process (Brownian motion)
Weak Convergence in Function Space
Ito Integral and Ito Calculus
Introduction to Diffusion Processes

STATISTICS 43800. Statistical Inference for Financial Data.
Sec 01: Per A. Mykland, TTh, 1:30-2:50 PM, Eckhart 117.
PQ: STAT 30200, 34300 and 38300 or consent of instructor.
Required Reading:

WINTER 2008

HSTD 43501. Theory and Methods for Multivariate and Longitudinal Data.
Sec 01: Paul Rathouz and Tyler VanderWeele, MW 1:30-2:50 pm, BSLC Arr.
PQ: Statistics 304, 301, 302, 343, 347. Statistics Statistics 244, 245, 246 with experience or coursework in matrix linear algebra may be substituted for Statistics 304, 301, 302.
Required reading: Instructor’s notes and selected papers.

This course presents a theoretical treatment of methods for multivariate and longitudinal data. The course covers both continuous and categorical data. Focus will be primarily on likelihood-based methods and their direct extensions. The first two-thirds of the course will cover mean and covariance models for multivariate normal data. Applications and special cases will include Hotelling's T-test,
multivariate linear regression, linear mixed and growth curve models, linear structural equations models and graphical models. The last one-third of the course will focus on categorical outcomes and will include generalized linear mixed models, structural equations models for categorical data and generalized linear marginal models. Readings will be taken from selected texts and original articles in the statistical literature. Students should expect four homework sets focused on theory and programming tasks related to methods developed in the course, as well as a final programming project.

Topic List (number of lectures) [expected instructor]
----------
(1) Review of Theory for Multivariate Normal Distribution [tv]
(1) Wishart Distribution and Hotelling's T-test [tv]
(4) General Linear Model for Correlated Data [pr]

Models for the mean and variance-covariance of multivariate processes
Likelihood methods for multivariate linear regression models
Hypothesis tests for variance-covariance parameters in non-standard situations

(7) Applications and Special Cases of the General Linear Model for Correlated Data [tv]

Linear Mixed Models and Growth Curve Models
Structural Equations Models
Multivariate Normal Graphical Models

(6) Generalized Linear Mixed Models -- Likelihood Methods [pr]

Canonical link models and the EM algorithm
Numerical integration techniques
MC integration

(time permitting) Structural Equations Models for Categorical Data [tv]

(time permitting) Generalized linear marginal models for categorical data (see Section 8.2 in DHLZ, Zhao and Prentice, Heagerty) [pr]

College Courses

STATISTICS 20000. Elementary Statistics.
Sec 01: Wei Biao Wu, MWF 9:30-10:20 AM, Eckhart 133.
PQ: Math 10600 or equivalent.
Required reading: Freedman, Pisani, and Purves, Statistics, 4th edition. W. W. Norton Press. ISBN-10: 0393929728, ISBN-13: 978-0393929720.

STATISTICS 22000. Statistical Methods and Their Applications.
Sec 01: Debashis Mondal, MWF 10:30-11:20 AM, Eckhart 133.
PQ: 2 QTRS Calculus.
Required reading: Moore and McCabe, Introduction to the Practice of Statistics, 5th edition. W. H. Freeman. ISBN-10: 0716764008, ISBN-13: 978-0716764007.

STATISTICS 22600=HSTD 32600. Analysis of Categorical Data.
Sec 01: Mei Wang, TTh, 3:00-4:20 PM, Eckhart 133.
PQ: STAT 22000 or equivalent.
Required reading: Agresti, A. An introduction to Categorical Data Analysis. Wiley, 2nd ed., 2007.

It is expected that the students have a good understanding of basic descriptive statistics such as means, variances and expectation, of the inferential notions of estimate, confidence intervals and significance or hypothesis testing. Familiarity with one statistical package, e.g. R, Splus, SAS, SPSS, Stata or Minitab, and ability to access Web sites and to download files from the Web are required. The free statistical package R will be used in this course for Winter 2007.

This course is an introduction to the theory and applications of statistical methods for investigating the relationships among discrete variables. The course will present methods for analyzing categorical data, including standard methods for contingency tables such as odds ratios,
tests of independence and various measures of association, generalized linear models for binary data and count data, logistic regression for binomial data, loglinear models for Poisson data, and models for paired samples with categorical responses. The statistical techniques discussed will be presented by many real examples involving physical, biological and social science data.

STATISTICS 22700=HSTD32700. Biostatistical Methods.
Sec 01: Ronald A. Thisted, TTh, 10:30-11:50 AM, BSLC 202.
PQ: HSTD 32400/STAT 22400 or STAT 24500 or equivalent; or consent of instructor.
Required reading: Collett, D. (2003). Modelling Binary Data, Second Edition. Boca Raton, Chapman & Hall/CRC.
Collett, D. (2004). Modelling Survival Data in Medical Research, Second Edition. London, Chapman & Hall.

This course is designed to provide students with tools for analyzing categorical, count, and time-to-event data frequently encountered in medicine, public health and related biological and social sciences. The course will emphasize application of methods rather than statistical theory, including
recognition of the appropriate methods, interpretation and presentation of results. Methods covered include: contingency table analysis, logistic regression, log-linear (Poisson) regression, conditional logistic regression, regression methods for ordinal data, Kaplan-Meier survival curves, parametric
survival models, and Cox proportional-hazards survival analysis.

STATISTICS 23400. Stati stical Models and Methods.
Sec 01: Linda B. Collins, MWF, 11:30-12:20 PM, Eckhart 133.
Sec 02: Nina Singhal Hinrichs, MWF, 2:30-3:20 PM, Ryerson 276 (as of 1/9/08).
PQ: Mathematics 13300, 15300 or 16300.
Required Reading: Chance and Rossman (2005). Investigating Statistical Concepts, Applications, and Methods, First Edition. Duxbury (Thomson Brooks/Cole), ISBN: 0-4950-5064-4.
Tanis and Hogg (2008). A Brief Course in Mathematical Statistics, Pearson/Prentice Hall, ISBN: 0-1317-5139-5.

STATISTICS 24400. Statistical Theory and Methods I.
Sec 01: Mathias Drton, TTh, 10:30-11:50 AM, Eckhart 133.
PQ: MATH 19600, 20100, or 20500.
Required reading: Rice, John A. (2007). Mathematical Statistics and Data Analysis, Third Edition, by (Duxbury).

This is the first quarter of a two-quarter sequence. Enrollment in the first quarter alone is permitted, although not recommended. The first quarter will cover the basics -- tools from probability and the elements of statistical theory. Topics will include the definitions of probability and random variables, binomial and other discrete probability distributions, normal and other continuous probability distribution, joint probability distributions and the transformation of random variables, principles of inference (including Bayesian inference), maximum likelihood estimation, hypothesis testing and confidence intervals, likelihood ratio tests, multinomial distributions and chi-square tests. Some large sample theory will be included. The emphasis will be upon statistical theory, specifically upon concepts and tools that are useful for understanding and applying statistical methodology.

There is no enforced prerequisite in probability or statistics, although the pace is such that students may find it useful to have taken a previous elementary course. The coverage of topics in probability will be limited and brief, so that those who have taken a course in probability will find reinforcement rather than redundancy. The second quarter will cover statistical methodology, including some multivariate analysis, the analysis of variance, the regression phenomenon, linear regression
analysis, data analysis, and correlation. Statistical software will be used for simulations and data analysis.

STATISTICS 24500. Statistical Theory and Methods II.
Sec 01: Wei Biao Wu, TTh, 1:30-2:50 PM, Eckhart 133.
PQ: STAT 24400 or consent of instructor.
Required reading: Rice, J. A. (2006). Mathematical Statistics and Data Analysis, 3rd ed., Brooks/Cole.

STATISTICS 24700=CPNS 32100. Math/Stats Methods for Neuroscience-2.
Sec 01: William Van Drongelen, WF, 1:30-2:50 PM, BSLC 401.
PQ: STAT 24400 or consent of instructor.
Required reading: Rice, J. A. (1995). Mathematical Statistics and Data Analysis, 2nd ed., Duxbury.

This course deals with the application of non-linear methods in signal processing and dynamical systems theory to issues in neuroscience. Data analysis with Matlab is again emphasized.

The third course in this sequence is an elective course in one of the quantitative sciences relevant to neuroscience that can be selected by the student in consultation with the program chair.

STATISTICS 25200=STAT 31200. Introduction to Stochastic Processes I.
Sec 01: Steven P. Lalley, TTh, 10:30-11:50 AM, Ryerson 277 .
PQ: STAT 25100 or consent of instructor
Required reading: Ross, S. (1996). Stochastic Processes, 2nd ed., Wiley.

Stochastic processes provide models for random events that evolve in time and may include substantial dependence among observations at different times. The goal of this course is to present a variety of useful models including Markov chains, renewal theory, random walks, queueing and branching processes.

Graduate Courses

STATISTICS 30100. Mathematical Statistics I.
Sec 01: Mary Sara McPeek, TTh, 1:30-2:50 PM, Eckhart 117.
PQ: STAT 30400 and MATH 20500 or consent of instructor.
Required Reading: Casella and Berger. Statistical Inference, 2nd ed.
Ferguson. A Course in Large Sample Theory.

This course is part of a two-quarter sequence on the theory of statistics. Topics will include exponential families, quadratic forms of multivariate normal, asymptotics of order statistics, sufficiency
and completeness, the likelihood function, methods of point estimation, and asymptotic properties of maximum likelihood estimates. Other topics (e.g. Bayesian methods and methods for dependent observations) may be covered if time permits.

STATISTICS 30600. Adv. Statistical Inference 1.
Sec 01: Peter McCullagh, TTh, 10:30-11:50 AM, Eckhart 117.
PQ: Consent of instructor.
Suggested reading: Tensor Methods in Statistics by P. McCullagh.
Principles of Statistical Inference by L. Pace and A. Salvan.
Statistical Models by A.C. Davison.

The focus of the course will be on parametric statistical models and likelihood.

Topics for discussion include

Statistical models: definition by example
Likelihood and likelihood ratio statistic
Asymptotic approximation of distributions
Edgeworth and related approximations
Cumulant calculations and tensor calculus
Marginal likelihood and REML estimation
Projection and Best linear prediction: splines
Processes, exchangeable processes, modulated processes, regression processes
Bayesian models

STATISTICS 31200=STAT 25200. Introduction to Stochastic Processes 1.
Sec 01: Steven P. Lalley, TTh, 10:30-11:50 AM, Ryerson 277.
PQ: STAT 25100 or consent of instructor.
Required Reading: Ross, S. (1996). Stochastic Processes, 2nd ed., Wiley.

STATISTICS 31700=STAT 25300. Introduction to Probability Models.
Sec 01: Mei Wang, TTh, 9:00-10:20 AM, Eckhart 133.
PQ: STAT 25100 or STAT 24400 or equivalent. Consent of instructor.
Required reading: Ross, R., Introduction to Probability Models, 9th ed., (2007).

This course introduces stochastic processes as models for a variety of phenomena in the physical and biological sciences. Another appropriate title for the course could be "an Introduction to Applied Stochastic Processes." Following a very brief review of basic concepts in probability the course will introduce stochastic processes that are popular in applications in sciences, such as discrete time Markov chain, the Poisson process, continuous time Markov process, renewal process and Brownian motion.

STATISTICS 32500=GSBC 41902. Statistical Inference.
Sec 01: Nicholas G. Polson, Tue 8:30-11:30 AM, GSB C10.
PQ: Business 41901=STAT 32500
Required reading: DeGroot and Scherviah, Probability and Statistics. Lecture notes will be provided in the form of a CoursePack.

This Ph.D.-level course is the second in a two-quarter sequence with Business 41901. The central topic is statistical inference. The topics covered include Bayesian inference, classical estimation, decision theory, MCMC methods and Hierarchical models. The use of Hierarchical models is a focus in applications.

STATISTICS 34500. Design and Analysis of Experiments.
Sec 01: Michael L. Stein, MW, 1:30-2:50 PM, Eckhart 133.
PQ: STAT 34300
Reading: Mead, R., The Design of Experiments. Cambridge.
West, B.D., Welch, K.B. and Galecki, A.T., Linear Mixed Models. Chapman & Hall/CRC.

An introduction to the methodology and application of linear models in experimental design. A major
focus of the course will be the basic principles of experimental design, such as blocking, randomization and incomplete layouts. Both standard designs, such as fractional factorials
and incomplete block designs, as well as nonstandard designs, will be studied within this context. The analysis of these experiments will be developed as well, with particular emphasis on careful model formulation and the role of fixed and random effects. Time permitting, additional topics may include the use of covariates in the analysis of designed experiments, spatial analysis of field trials and Bayesian approaches to analysis of experimental data.

Course work will include the planning, execution and analysis of an experiment by the class.

STATISTICS 35540. Population Genetics and Metagenomics.
Sec 01: Nicholas Eriksson, MW, 1:30-2:50 PM, Eckhart 117.
Required reading: Selected papers.

In this course we will focus on the biological, statistical, and computational challenges involved in the analysis of metagenomic data. In metagenomics and population genetics, the goal is to understand the patterns of variation in genomes within species and between related species. New sequencing
technologies allow us to gather data on unexplored populations, but require new methods of analysis. We will look at the connections between these problems and more established fields such as phylogenetics, sequence assembly, and population genetics.

Topics will be chosen from recent papers from the literature; some of these papers will be presented by seminar participants.

STATISTICS 35600=HSTD 33100. Applied Survival Analysis.
Sec 01: James Dignam, TTh, 10:30-11:50 AM, BSLC 305.
PQ: HSTD 32100; STAT 22000; or equivalent, and HSTD 32400/STAT 22400 or equivalent; or consent of instructor.
Required reading:

This course will provide an introduction to the principles and methods for the analysis of time-to-event data. This type of data occurs extensively in both observational and experimental biomedical and public health studies, as well as in industrial applications. While some theoretical statistical detail is given (at the level appropriate for a Master's student in statistics), the primary focus will be on data analysis. Problems will be motivated from an epidemiologic and clinical perspective, concentrating on the analysis of cohort data and time-to-event data from controlled clinical trials.

STATISTICS 35700=HSTD 31001. Epidemiologic Methods.
Sec 01: Diane Lauderdale, TTh, 12:00-1:20 PM, BH W229.
PQ: HSTD 30900 or consent of instructor.
Required reading:

This course expands on the material presented in "Principles of Epidemiology," further exploring issues in the conduct of epidemiologic studies. The student will learn the application of both stratified and multivariate methods to the analysis of epidemiologic data. The final project will be to write the "specific aims" and "methods" sections of a research proposal on a topic of the student's choice.

STATISTICS 37300. Graphical Models and Algebraic Statistics.
Sec 01: Mathias Drton, TTh, 1:30-2:50 PM, Eckhart 202.
PQ: Consent of instructor.
Required reading: None.

In graphical modelling, one associates a statistical model with a graph: nodes represent variables and edges indicate dependencies between variables. This framework is popular in many applied areas and encompasses models such as Bayesian networks, log-linear models, phylogenetic tree models, multivariate regression, factor analysis, and structural equation models. Topics to be discussed in this course include conditional independence, directed and undirected graphical models, hidden variables, and statistical inference in the different model classes. We will also discuss algebraic techniques that are useful in the study of graphical models.

STATISTICS 38300. Measure-Theoretic Probability-3.
Sec 01: Michael J. Wichura, MWF, 12:30-1:20 PM, Eckhart 117.
PQ: STAT 38100 or consent of instructor.
Required reading: No text book is required; notes will be distributed in class.

Topics for Stat 38300 will include:

The Hahn and Jordan decomposition theorems
Modes of convergence: with probability one, in probability, and in mean; uniform integrability
L2-spaces: projections; representation of linear functionals
The Radon-Nikodym theorem: absolute continuity, Radon-Nikodym derivatives; likelihood ratios; Lebesgue decompositions
Conditional probability: regular conditional probability distributions
Conditional expectation: given sub-sigma fields, and given measurable functions
Martingales: definitions and examples, transformations
Stopping times; optional sampling
Martingale limit and closure theorems
Backward submartingales
Continuous-time martingales: convergence, closure, optional sampling

Statistics 39000=FINM 34500. Stochastic Calculus-1
Sec 01: Jostein Paulsen, W, 6:00-9:00 PM, Ryerson 251.
PQ: Math Finance Students Only. The course, Mathematical Foundations of Option Pricing is very useful. Otherwise, the more mathematical analysis you know, the better.
Required Reading: Tomas Bjork: Arbitrage Theory in Continuous Time. Oxford University Press
Steven Shreve: Stochastic Calculus for Finance II: Continuous-Time Models. Springer Finance.

In the class we will give a more or less in depth coverage of the following topics. Notes are provided by the instructor.

• Basic stochastic calculus including martingales, filtrations, Brownian motion, the
Itˆo integral, Itˆo’s formula and stochastic differential equations.

Option pricing, both with deterministic interest rate and with stochastic interest rate. Also option hedging.
Forwards and futures
Foreign exchange derivative pricing
Interest rate models including classical models, the Heath-Jarrow-Morton approach and the Market Models.
Some credit risk models

STATISTICS 47900. Stochastic Models for Memory and Learning.
Sec 01: Yali Amit , TTh, 3:00-4:20 PM, Eckhart 117. Course begins 6th week.
PQ: Consent of instructor.
Required reading: None

This 5 week course will cover a some of the literature analyzing learning and memory in large neuronal populations as stochastic processes. First we will discuss models with discrete time dynamics, discrete binary neurons and finite state synapses, and derive bounds on memory capacity, learning and forgetting times. This will only involve discrete time Markov chain analysis and some ideas from mean-field analysis. Second we will introduce continuous integrate and fire neurons and continuous time dynamics. Using mean-field methods we will analyze the stability of large networks with random connections, and the behavior of the networks after learning. People interested in probability will be exposed to a rich collection of stochastic models waiting to be analyzed in a rigorous mathematical framework (the mean field analysis is only approximate.) People interested in neuroscience will be exposed to interesting models and some intriguing connections to experimental
data.

STATISTICS 48400. Adv. Topics in Probability 1.
Sec 01: Steven P. Lalley, TTh, 3:00-4:20 PM, Eckhart 117. Course held 1st-5th weeks of quarter.
PQ: Basic probability and linear algebra, some elementary complex variable theory.
Recommended reading:

"Methodologies in spectral analysis of large dimensional random matrices, a review" by Z. D. Bai , Statistica Sinica 9 (1999), 611-677.
"High dimensional statistical inference and random matrices" by I. Johnstone, http://front.math.ucdavis.edu/0611.5589.
"Introduction to the Random Matrix Theory: Gaussian Unitary Ensemble and Beyond" by Yan V. Fyodorov, http://front.math.ucdavis.edu/0412.4717.
"Orthogonal Polynomials and Random Matrices: A Riemann-Hilbert Approach" by P. Deift (mainly ch. 5), http://www.ams.org/bookstore.

This 5-week course will introduce the basic theory of large random matrices. The first part of the course will be devoted to "bulk spectral" properties of Wigner and sample covariance matrices
(that is, the empirical distribution of their eigenvalues), leading to the Wigner semi-circle law and the Marchenko-Pastur theorem. The second part will focus on the top of the spectrum, and will (I hope) give a mostly complete derivation of the "Tracy-Widom" distribution.

AUTUMN 2007

College Courses

STATISTICS 20000. Elementary Statistics
Sec 01: Xinghua Zheng, MWF 9:30-10:20 AM, Eckhart 133
Sec 02: Minsun Song, MWF 12:30-1:20 PM, Eckhart 133
PQ: Math 10600 or equivalent.
Required reading: Freedman, Pisani, and Purves, Statistics, 4th edition. W. W. Norton Press. ISBN-10: 0393929728, ISBN-13: 978-0393929720.

STATISTICS 22000. Statistical Methods and Their Applications
Sec 01: Shali Wu, MWF 10:30-11:20 AM, Eckhart 133
Sec 02: Marcin Hitczenko, MWF 1:30-2:20 PM, Eckhart 133
PQ: 2 QTRS Calculus.
Required reading: Moore and McCabe, Introduction to the Practice of Statistics, 5th edition. W. H. Freeman. ISBN-10: 0716764008, ISBN-13: 978-0716764007.

STATISTICS 22400=HSTD 32400. Applied Regression Analysis
Sec 01: Vanja Dukic, TTh 10:30-11:50 AM, Eckhart 133
PQ: HSTD 32700 or STAT 22000 or STAT 23400 or STAT 24400 or consent of instructor.
Required reading:

This course is an introduction to the methods and applications of fitting and interpreting multiple regression models. The main emphasis is on the method of least squares. Topics include the examination of residuals, the transformation of data, strategies and criteria for the selection of a regression equation, the use of dummy variables, and tests of fit. The techniques discussed will be illustrated by many real examples involving biological and social science data. Examples and exercises will be implemented in a statistical software package "Stata", but familiarity with Stata is not required.

STATISTICS 23400. Statistical Models/Method-1
Sec 01: Linda Collins, MWF 11:30-12:20 PM, Eckhart 133
Sec 02: Nicholas Eriksson, MWF 2:30-3:20 PM, Eckhart 133
PQ: MATH 13300, 15300 or 16300
Required reading: Chance and Rossman, Investigating Statistical Concepts, Applications, and Methods. Thompson, Brooks/Cole, ISBN-10: 0495050644, ISBN-13: 978-0495050643.

This course presents basic ideas of probability theory and statistics, and is recommended for students throughout the natural and social sciences who want a broad background in statistical methodology and exposure to probability models and the statistical concepts underlying the methodology. Probability is developed for the purpose of modeling outcomes of random phenomena. Random variables and their expectations are studied; including means and variances of linear combinations, and an introduction to conditional expectation. Binomial, Poisson, normal and other standard probability distributions are considered. Some probability models are studied mathematically and others via simulation on a computer. Sampling distributions and related statistical methods are explored mathematically, studied via simulation and illustrated on data. Statistical methods for describing data and making inferences based on samples from populations are presented. Methods include, but are not limited to, inference for means and variances for one- and two-sample problems, correlation and simple linear regression. Graphical and numerical data description are used for exploration, communication of results, and comparing mathematical consequences of probability models and data. Mathematics is employed to the level of univariate calculus and is less demanding than that required by STAT 24400.

STATISTICS 24400. Statistical Theory/Method-1
Sec 01: Michael Stein, TuTh, 1:30-2:50 PM, Eckhart 133
PQ: MATH 19600, 20100, or 20500
Required Reading: Rice, John A. (2007). Mathematical Statistics and Data Analysis, Third Edition, by (Duxbury).

This is the first quarter of a two-quarter sequence. Enrollment in the first quarter alone is permitted, although not recommended. The first quarter will cover the essential tools from probability needed for study of statistical theory and the basic elements of statistical theory.

Topics will include the definitions of probability and random variables, binomial and other discrete probability distributions, normal and other continuous probability distribution, joint probability distributions and the transformation of random variables, principles of inference (including Bayesian inference), maximum likelihood estimation, hypothesis testing and confidence intervals, likelihood ratio tests, multinomial distributions and chi-square tests. Some large sample theory will be included. The emphasis will be upon statistical theory, specifically upon concepts and tools that are useful for understanding and applying statistical methodology.

There is no enforced prerequisite in probability or statistics, although the pace is such that students may find it useful to have taken a previous elementary course. The coverage of topics in probability will be limited, so that those who have taken a course in probability will find reinforcement rather than redundancy. The second quarter will cover statistical methodology, including some multivariate analysis, the analysis of variance, the regression phenomenon, linear regression analysis, data analysis, and correlation.

The mathematics prerequisites are listed as general guidance. You should be comfortable with multivariate calculus through partial differentiation and multiple integration.

Graduate Courses

STATISTICS 30400. Distribution Theory
Sec 01: Dan Nicolae, MWF, 1:30-2:20 PM, Eckhart 117
PQ: STAT 24500 or MATH 25000 or equivalent
Recommended reading: Severini, T. (2005). Elements of Distribution Theory. Cambridge University Press.

This course covers the basics of distribution theory. Topics include:

Distribution functions and their inverses, quantile functions, Q/Q plots, change-of-variables for probability densities
Expectation, variance, median, mode of random variables
Basics of measure theory, including Fubini's theorem and interchangeability of limits and integrals
Moment generating functions and characteristic functions, including power series expansion, inversion formulas, uniqueness theorems, and convergence in distribution
Cumulants and cumulant generating functions: examples, properties
Different concepts of convergence for random variables
Limit theorems, including the weak law of large numbers, the central limit theorem.

STATISTICS 30700=CMSC 37800. Numerical Computation
Sec 01: Ronald Thisted, TTh, 10:30-11:50 AM, Ryerson 276
PQ: Stat 34300 (concurrent enrollment OK) or consent of instructor.
Required reading: Thisted, Ronald A. Elements of Statistical Computation. CRC/Chapman & Hall.
Recommended, but not required:
   Gentle, James. Random Number Generation and Monte Carlo Methods. Second edition.
   Springer.
   Watkins, David S. Fundamentals of Matrix Computations. Second edition.
   Wiley.
   Scheinerman, Edward. C++ for Mathematicians. CRC/Chapman and Hall.

This course starts with a presentation of the fundamental algorithms for the solution of linear equations, the decomposition of matrices, and finite dimensional eigenvalue problems. Applications to least squares/regression will be presented, emphasizing use of existing numerical software. The course will also discuss optimization problems and introduce the basic principles of simulation-based methods.

Topics include:

Gaussian elimination and back-substitution
LU decomposition. (General/Symmetric)
Singular value decomposition. (Symmetric)
Householder orthogonalization and QR factorization. (Symmetric).
Iterative methods: Jacobi and Gauss Seidel.
Optimization: Newton-Raphson and quasi-Newton.
Uniform random number generation.
Simulating specific distributions
Monte Carlo methods

By the end of the course students should be able to apply these algorithms in their research work.

STATISTICS 32301=HSTD 43001. Advanced Bayesian Methods.
Sec 01: Vanja Dukic, W, 12:30-3:20 PM, BH W230
PQ: STAT 32300/HSTD 43000 or STAT 30100-30200, 31200-31300 and consent of instructor
Required reading: Notes and manuscripts will be distributed in class.

This class is a continuation of the Bayesian Topics (Stat 32300/HSTD 43000). We will move beyond the material learned there (the basics of Bayesian statistics and computation (importance sampling, EM, MCEM, data augmentation, Metropolis-Hastings and Gibbs sampling). In particular, we will focus on extensions to MCMC geared for dealing with high-dimensional problems with potential multimodality (simulated tempering, sequential Monte Carlo, Hamiltonian MCMC, Langevin MCMC). We will also discuss issues and algorithms for model comparison (transdimensional MCMC and algorithms for computation of normalizing constants). Algorithms can be implemented in any language, but familiarity with R or Matlab will be assumed. The class will have a seminar format.

STATISTICS 32400=GSBC 41901. Probability and Statistics.
Sec 01: Nicholas Polson, Tue, 8:30-11:30 AM, HC3B
PQ: One year of Calculus
Required reading: The text for the course is DeGroot and Schervish, Probability and Statistics. Lecture notes will be available in the form of a CoursePack.

STATISTICS 33100. Sample Surveys
Sect 01: Kirk Wolter, TTh, 10:30-11:50 AM, Eckhart 117
PQ: Consent of instructor
Required reading: Wolter, K.M. (2007). Introduction to Variance Estimation, 2nd Edition, Springer-Verlag, New York.

This is an introductory course to the statistics and methodology of sample surveys. Topics include

basic methods of sample selection,
determining sample size, stratification,
general estimators (Horvitz-Thompson, ratio, generalized regression, calibration)
domain estimation,
nonresponse,
nonsampling error,
multiple-stage sampling,
a national sampling frame for area probability surveys,
telephone surveys,
questionnaire design,
variance estimation for complex surveys,
analysis of contingency tables, and
regression analysis for survey data.

The course will be of interest to students who anticipate a research career that designs, collects, and analyzes survey data in fields such as economics, education, healthcare, marketing, psychology, sociology, and statistics.

STATISTICS 33900=FINM 33100. Financial Data Analysis
Sec 01: Per A. Mykland, W, 6:00-9:00 PM, Eckhart 202
PQ: Math Finance Students only
Required reading:
Recommended textbooks:
Applied Linear Regression (3rd edition) by Sanford Weisberg (Wiley)
Time Series: Applications to Finance by Ngai Hang Chan (Wiley)
Statistical Analysis of Financial Data in S-Plus by Rene A. Carmona (Springer-Verlag)
Mathematical Statistics and Data Analysis by John A. Rice (Duxbury)
The Basics of S-plus, by Andreas Krause and Melvin Olson (Springer-Verlag)

Note that only part of each book will be used in the course. However, in the course of your career, you will find all of them useful to have on your shelf for further reading and reference.

STATISTICS 34300. Applied Linear Stat Methods
Sec 01: Mathias Drton, TTh, 9:00-10:20 AM, Eckhart 133
PQ: STAT 24500 and MATH 25000 or equivalent
Optional Reading: Venables, W.N. and Ripley, B.D. (1999). Modern Applied Statistics with S-Plus (3rd ed). Springer-Verlag.
Required Reading: Weisberg, S. (2005). Applied Linear Regression, Third Edition. John Wiley & Sons. Software: Splus or R.

Statistics 34300 is an intensive course in the theory and methods of linear regression and related techniques of statistical modelling. It is intended primarily for graduate students in Statistics and related fields.

The course is also open to undergraduates and others who have a solid understanding of matrix algebra and basic statistical theory. Thorough familiarity with the simple linear regression model is expected.

The course will review linear regression with a single predictor, and will cover the multiple-predictor case; least-squares estimation; associated distribution theory; estimation, confidence intervals and tests; regression with errors in the predictors; weighted least squares, assessing lack of fit; residual analysis; regression diagnostics; transformations; model building; collinearity; subset-selection methods, including stepwise regression; prediction; nonlinear least squares.

STATISTICS 35000=HSTD 33300, ENST 27400, PPHA 36400. Principles of Epidemiology
Sec 01: Kurina Lianne , TTh, 9:00-10:20 AM, BSLC 240
PQ: Introductory Statistics
Required reading:

STATISTICS 35600. CANCELLED
Sec 01

Back to to

STATISTICS 36900=HSTD 33300. Longitudinal Data Analysis
Sec 01: Paul Rathouz, TTh, 9:00-10:20 AM, BSLC 313
PQ: HSTD 32100; STAT 22000 or equivalent and HSTD 32400-STAT 22400 or equivalent; or consent of instructor.
Required reading: *Fitzmaurice GM, Laird NM, Ware JH. (2004). Applied Longitudinal Analysis. Hoboken, NJ: Wiley.
Other references: (* Indicates text on reserve in Eckhart Library.)
   *Diggle PJ, Heagerty P, Liang K-Y, & Zeger SL. (2002). Analysis of Longitudinal Data, 2nd edn.
Oxford: Oxford University Press.
   *Hedeker DR & Gibbons RD. (2006). Longitudinal data analysis. Hoboken, NJ: Wiley-Interscience.
Hand D, Crowder M. (1996). Practical Longitudinal Data Analysis. London: Chapman & Hall.
   *Lindsey JK. (1999). Models for repeated measurements. New York: Oxford University Press.
Littel RC, Milliken GA, Stroup WA, Wolfinger RD. (1996). SAS System for Mixed Models. Cary,
NC: SAS Institute.

STATISTICS 37610. Monte Carlo Methods in Scientific Computing
Sec 01: Samuel Kou, TTh, 1:30-2:50 PM, Eckhart 117
PQ: Consent of instructor
Required reading: Jun S. Liu (2001). Monte Carlo Strategies in Scientific Computing. Springer.

This course introduces students to modern Monte Carlo methods and their applications in scientific computing. In addition to the classical topics, such as Metropolis-Hastings algorithm, Gibbs sampler, importance sampling and sequential Monte Carlo, special topics include Swendsen-Wang algorithm, multicanonical sampling, histogram method, multigrid sampling, parallel tempering, equi-energy sampler, fragment-regrowth sequential sampling, dimension expansion Monte Carlo, DNA sequence alignment and motif finding, lattice protein folding, and polymer models on lattice.

STATISTICS 37910=CMSC 35510. Statistical Methods in Computer Vision
Sec 01: Yali Amit, TTh, 10:30-11:50 AM, Ryerson 277
PQ: Consent of instructor
Required reading: Instructor’s notes and selected papers.

The first part of this course will provide an introduction to pattern recognition, and an overview of statistical modeling of high dimensional data and parameter estimation methods. Sub jects include graphical models, mixture models, Maximum likelihood and Bayesian estimation, the EM algorithm and its variants. These methods will then be employed in a number of computer vision algorithms involving ob ject detection and recognition. Class work will consist of some theoretical assignments and implementation of some of the algorithms in code on real data.

STATISTICS 38100. Measure-Theoretic Probability I
Sec 01: Michael Wichura, MWF, 2:30-3:20 PM, Eckhart 117
PQ: STAT 31300 or consent of instructor
Required Reading: There is no text required or recommended. Notes will be provided.

This course is the first of a three quarter sequence presenting a careful development of some topics from measure and probability. Topics to be covered in 381 include: classes of sets -- fields, sigmafields, monotone classes, pi and lambda systems; probabilities and general measures; independence and the Borel-Cantelli lemmas; measurable functions; induced measures, distribution and inverse distribution functions; integration with respect to measures -- basic properties, change of variable, indefinite integration, densities; integration to the limit -- MCT, DCT, and friends; laws of large numbers, applications to probability and statistics; transition probabilities and product measures.

SUMMER 2007

STATISTICS 22000. Stat Meth And Applications
Sec 01: David Matteson, TTh, 1:00-3:00 PM, Eckhart 117
PQ: Math 15200 or equivalent
Required reading: Moore, D. S. and McCabe, G. P. (2006). Introduction to the Practice of Statistics, 5th Edition. Freeman.

This course is an introduction to statistical techniques and methods of data analysis, including the use of computers. Examples are drawn from the biological, physical, and social sciences. Students are required to apply the techniques discussed to data drawn from actual research. Topics include data description, graphical techniques, exploratory data analyses, random variation and sampling, one- and two-sample problems, the analysis of variance, linear regression, and analysis of discrete data.

SPRING 2007

STATISTICS 22200. Linear Models and Experimental Design.
Instructor: Mei Wang.

Time: TTh, 9:00-10:20 AM.
Location: Eckhart 133.
PQ: STAT 22000 or consent of instructor.
Required Reading: Oehlert, G. W. (2000) A First Course in Design and Analysis of Experiments. W. H. Freeman. ISBN: 0-7167-3510-5.

This course will introduce the student to the major statistical issues in the design of experiments and the analysis of experimental data. The major topics will be

The basic principles of experimental design: randomization, blocking,
and balance.
Important classes of designs: matched pairs, complete factorial designs,
and fractional factorial designs.
Analysis of variance and inference from experimental data.

In addition to regular homework assignments and exams, students will be required to complete a project involving the design and analysis of an experiment of their own.

STATISTICS 23400. Statistical Models and Methods.
Instructor: Linda B. Collins (Section 01). Oli Atlason (Section 02).

Time: MWF, 11:30-12:20 PM (Section 01). MWF, 2:30-3:20 PM (Section 02).
Location: Eckhart 133.
PQ: Math 13300, 15300 or 16300.
Required Reading: Chance and Rossman (2005). Investigating Statistical Concepts, Applications, and Methods, First Edition. Duxbury (Thomson Brooks/Cole), ISBN: 0-4950-5064-4.

This course presents basic ideas of probability theory and statistics, and is recommended for students throughout the natural and social sciences who want a broad background in statistical methodology and exposure to probability models and the statistical concepts underlying the methodology. Probability is developed for the purpose of modeling outcomes of random phenomena. Random variables and their expectations are studied; including means and variances of linear combinations, and an introduction to conditional expectation. Binomial, Poisson, normal and other standard probability distributions are considered. Some probability models are studied mathematically and others via simulation on a computer. Sampling distributions and related statistical methods are explored mathematically, studied via simulation and illustrated on data. Statistical methods for describing data and making inferences based on samples from populations are presented. Methods include, but are not limited to, inference for means and variances for one- and two-sample problems, correlation and simple linear regression. Graphical and numerical data description are used for exploration, communication of results, and comparing mathematical consequences of probability models and data. Mathematics is employed to the level of univariate calculus and is less demanding than that required by STAT 24400.

STATISTICS 24500. Statistical Theory/Method-2.
Instructor: Stephen M. Stigler.

Time: TTh, 1:30-2:50 PM.
Location: Eckhart 133.
PQ: STAT 23400 and STAT 23500 or STAT 24400 or consent of instructor.
Required Reading: Rice, J. A. Mathematical Statistics and Data Analysis, 3rd ed., Duxbury Press. ISBN: 0-534-20934-3.

STATISTICS 24600. Statistics Theory/Method-3.
Instructor: Yali Amit.

Time: TTh, 10:30-11:50 AM.
Location: Eckhart 133.
PQ: STAT 24400 and STAT 24500 or consent of instructor.
Required Reading: Wasserman, Larry (2003). All of Statistics, Springer. ISBN: 0-387-40272-1.

This course will introduce a variety of modern statistical methods. We will start with the description of families of multivariate models (multivariate normal distribution, graphical models, log-linear models). We will then discuss Missing data problems and the EM algorithm, Bayesian inference, Monte-carlo simulations, bootstrap and some modern classification techniques.

STATISTICS 25100. Intro to Math Probability.
Instructors: Michael J. Wichura and Gregory F. Lawler. Professor Wichura will teach both sections 01 and 02 for the first half of the course; Professor Lawler will teach both sections
the second half of the course.

Time: TTh, 12:00-1:20 PM (Section 01). TTh, 3:00-4:20 PM (Section 02).
Location: Eckhart 312.
PQ: MATH 20000 or MATH 20500 or consent of instructor.
Required Reading: Ross, A First Course in Probability, 7th Edition. Pearson/Prentice Hall. ISBN 0-13-185662-6.

The aim of the course is to provide an introduction to the concepts of probability. The course will cover the basic ideas used to describe aspects of randomness, such as events, random variables, independence, and conditional probability, with emphasis on the methods, calculation, and applications of probability. The topics treated are: combinatorics; probability models; rules of probability; conditional probability; independence; random variables; expectation and standard deviation; games of chance; common discrete distributions --- uniform,
binomial, hypergeometric, geometric, negative binomial, and Poisson --- and their interrelationships; univariate and multivariate density and distribution functions, change of variable formulas for densities, common continuous distributions --- beta, Cauchy, chisquare, exponential, gamma, lognormal, normal, uniform --- and their interrelationships; moment generating functions, laws of large numbers and the central limit theorem.

Midterm: Please note that there will be a COMMON midterm for both sections, to be held THURSDAY EVENING of the sixth week (May 3), from 6:00-8:00PM

For more information, please visit http://galton.uchicago.edu/~wichura/Stat251/courseinfo.html

STATISTICS 26700=STAT 36700=HIPS 25600=CHSS 32900. History of Statistics.
Instructor: Stephen M. Stigler.

Time: MWF, 9:30-10:20 AM.
Location: Eckhart 203.
PQ: A course in Statistics.
Recommended Reading: Stephen M. Stigler, The History of Statistics: The Measurement of Uncertainty Before 1900. (Cambridge, Mass.: Harvard University Press, 1986.) (Available in paperback.) Other materials will be distributed in class or by web.

STATISTICS 30200. Mathematical Statistics-2.
Instructor: Wei-Biao Wu.

Time: MW, 1:30-2:50 PM.
Location: Eckhart 117.
PQ: STAT 30100 or consent of instructor.
Required Reading: Casella and Berger. Statistical Inference, Second edition. Duxbury Press. ISBN: 0-534-24312-6

STATISTICS 31300. Intro: Stochastic Processes-2.
Instructor: Steven P. Lalley.

Time: TTh, 10:30-11:50 AM.
Location: Eckhart 117.
PQ: STAT 31200 or consent of instructor.
Required Reading: Notes will be posted online.

This course is a continuation of Statistics 31300: Introduction to Stochastic Processes I. Topics to be discussed will include rates of convergence for Markov chains, introduction to MCMC, generating functions, Galton-Watson processes, continuous-time Markov chains, martingales, and Brownian motion. There will be weekly homework assignments, and midterm and final exams.

STATISTICS 32300. Bayesian Methods in Biostatistics.
Instructor: Vanja Dukic.

Time: Wed, 12:30-3:20 PM
Location: BSLC
PQ: STAT 30100-30200, 24400-24500, 34300, 31200-31300, consent of instructor.
Required Reading: Tanner, 3rd ed., Tools for Statistical Inference: Methods for the Exploration of Posterior Distributions and Likelihood Functions, Springer.

This course will cover basics of modern statistical computation, with emphasis on Bayesian computational methods. It will begin with the introduction to Bayesian statistics, and cover normal and non-normal approximation to likelihood and posterior distributions, the EM algorithm, data augmentation and Markov Chain Monte Carlo (MCMC) methods. Time permitting, we will conclude with some recent developments in the MCMC area, such as perfect and adaptive sampling methods. Biostatistical and environmental examples will be given throughout the course. There will be weekly homeworks, and students will be expected to complete a project by the end of the course. There will be no final exam, but there will be an in-class final project presentation. Algorithms can be implemented in any language, but familiarity with R and Matlab will be assumed.

STATISTICS 33200=HSTD 43200. Causal Inference.
Instructor: Tyler J. VanderWeele.

Time: MW, 10:30-11:50 AM
Location: BSLC
PQ: The course is intended for both masters and doctoral students in statistics and in the social sciences. The course will be accessible to anyone with a firm understanding of linear and logistic regression (e.g. STAT 224/226 or HSTD 324/327) though students would benefit from a more sophisticated understanding of statistics. Supplementary handouts will be provided covering the proofs and technical details concerning methods presented for more theoretically-oriented students.
Required Reading: None. If you want to read further on certain topics, take a look at Judea Pearl's book "Causality".

The course will be concerned with the process of drawing causal inferences from observational data in the biomedical and social sciences. The course will introduce a number of fundamental concepts in causal inference and cover methods of estimating causal effects for both point- and time-varying exposures. Concepts and methods that will be covered include: confounding, potential outcomes, propensity scores, directed acyclic graphs, inverse probability of treatment weighting, marginal structural models and structural nested models. Time permitting, the course may also briefly survey a number of other topics such as instrumental variables, the estimation of direct and indirect effects, sensitivity analysis and bounds for causal effects.

STATISTICS 33700=GSBC 41914=ECON 31500. Multivariate Time Series Anal.
Instructor: Ruey-Shiong Tsay.

Time: Wed, 8:30-11:30 AM
Location:
PQ: Business 41910 or equivalent; = STAT 33700 and ECON 31500
Required Reading: None.

Course website: http://faculty.chicagogsb.edu/ruey.tsay/teaching/mts

A useful web site for U.S. data: Fed. Res. at St Louis: http://research.stlouisfed.org/fred2/

Course Objective:
1. To study the basic theory of multivariate processes
2. To gain experience in analyzing multivariate time series data
3. To learn multivariate time series models, including vector AR and ARMA models with exogenous variables
4. To understand co-integration and error-correction models
5. To study structural specification of a vector process
6. To learn state-space models and Kalman filter.

No textbook is assigned. Lecture Notes available on course web.

Grading:
Mid-term (40%), Final project (40%), and homework assignments (20%).

Computing:
The key software packages are SCA and S-Plus, but you may use any software of your choice.

Course Outline: All topics include applications
1. Transfer function models
2. Stationary vector autoregressive and moving average models
3. Estimation, modeling, and forecasting
4. Unit-root, co-integration and error-correction models
5. Multivariate Seasonal models
6. Structural specification
7. State-space model and Kalman filter
8. Multivariate volatility models if time permits.

STATISTICS 34700. Generalized Linear Models.
Instructor: Peter McCullagh.

Time: TTh, 3:00-4:20 PM.
Location: Eckhart 133.
PQ: STAT 34300 or consent of instructor.
Required Reading: McCullagh & Nelder, Generalized Linear Models, 2nd edition. Chapman & Hall/CRC. ISBN: 0-412-31760-5.
Recommended Reading: Venables & Ripley, Modern Applied Statistics w/S, 4th edition, Springer. ISBN: 0-387-95457-0.
Cox & Snell (2000), Applied Statistics, Chapman & Hall/CRC. ISBN: 0-412-16570-8.

STATISTICS 35500. Statistical Genetics.
Instructor: Mary Sara McPeek.

Time: Friday, 1:30-4:00 PM.
Location: Eckhart 117.
PQ: Consent of instructor.
Required Reading: None.

This is an advanced course in statistical genetics. Prerequisites are Human Genetics 47100 and Statistics 24400 and 24500. Students who do not meet the prerequisites may enroll on a P/NP basis with consent of the instructor. This is a discussion course and student presentations will be required. Topics vary and may include, but are not limited to, statistical problems in association mapping, linkage mapping, population genetics, microarray analysis, and genetic models for complex traits.

STATISTICS 36700=STAT 26700=CHSS 32900=HIPS 25600. History of Statistics.
Instructor: Stephen M. Stigler.

STATISTICS 37800. Statistical Computing.
Instructor: Dongping Fang.

Time: TTh, 1:30-2:50 PM.
Location: Eckhart 117.
PQ: Experience with R or Splus programming.
Required Reading: None.

This course emphasizes practical aspects of designing statistical algorithms for implementation. Throughout the course, some basic computer arithmetic, accuracy, linear algebra and optimization methods are covered, some selected algorithms like linear regressions, multinomial logistic regression, neural networks, classification tree, clustering will also be introduced and used for illustration. Note that this course doesn't cover program languages like C++, Java, etc.

References:

Lange, K. (1999). Numerical Analysis for Statisticians. Springer.
Press, W. H., Teukolsky, S. A., Vetterling, W. T., and Flannery, B. P. (1992). Numerical Recipes in C: The Art of Scientific Computing. (or Numerical Recipes in Fortran: The Art of Scientific Computing.) Second Edition. Cambridge University Press.
Thisted, R. A. (1988). Elements of Statistical Computing. Chapman and Hall.
Gray, R. (2003), Advanced Statistical Computing (BIO 248 cd Course Notes, http://www.stat.wisc.edu/~mchung/teaching/stat471/stat_computing.pdf)
Micah Altman, M., Gill, J., and McDonald, M. P. (2004). Numerical Issues in Statistical Computing for the Social Scientist. John Wiley & Sons

STATISTICS 38600. Topics in Stochastic Processes.
Instructor: Steven P. Lalley.

Time: TTh, 3:00-4:20 PM
Location: Eckhart 117
PQ: STAT 23400 and STAT 23500 or STAT 24400 or consent of instructor.
Required Reading:

Topics:

Wiener process (Brownian motion)
Weak Convergence in Function Space
Ito Integral and Ito Calculus
Introduction to Diffusion Processes