Second semester of sequence in theory of linear models with focus on "analysis of variance/design of experiments" models. Includes factorial experiments (balanced and unbalanced designs, notions of interaction, etc.), randomized block designs, incomplete designs (incomplete block designs and latin squares), random effects, nested models, and mixed models.

Time series analysis is an effective statistical methodology for modelling time series data (a series of observations collected over time) and forecasting future observations in many areas; economics, the social sciences, the physical and environmental sciences, medicine, and signal processing. For example, monthly unemployment rates in economics, yearly birth rates in social science, global warming trends in environmental studies, and magnetic resonance imaging of brain waves in medicine.

This course focuses on modern statistical methods for analysis of network data, especially social network data. Networks are characterized by two types of units: Nodes (often representing people), and edges representing the relations between nodes. Network data are characterized by the nesting of edges between nodes, which creates interesting features for statistical dependence and sampling. In addition to statistical methods, the course will include some discussion of connections to motivating problems and underlying social science theory.

Contact department for description.

Students will listen to, and have a chance to participate in, weekly sessions on the experience of statistical consulting

Point and interval estimation, hypothesis testing, large sample results in estimation and testing; decision theory; Bayesian methods; analysis of discrete data. Also, topics from nonparametric methods, sequential methods, regression, analysis of variance. Prerequisite: Statistc 607 or equivalent.

A modern treatment of probability theory based on abstract measure and integration. Random variables, expectations, independence, laws of large numbers, central limit theorem, and general conditioning using the Radon-Nikodym theorem. Introduction to stochastic processes: martingales, Brownian motion. Prerequisite: Math 623.

This course aims to introduce basic concepts and modeling techniques for time series data. It emphasizes implementation of the modeling techniques and their practical application in analyzing actuarial and financial data. The open source program R will be used. Chapter 7, 8 and 9 of "Regression Modeling with Actuarial and Financial Applications", by E.W. Frees, Cambridge University Press, 2010 will be covered, if time allows.

This course provides a calculus-based introduction to probability and statistical inference. Topics include the axioms of probability, sample spaces, counting rules, conditional probability, independence, random variables and distributions, expected value, variance, covariance and correlation, the central limit theorem, random samples and sampling distributions, basic concepts of statistical inference (point estimation, confidence intervals and hypothesis testing) and their use in one and two-sample problems.