New Fall Course Offering: CSI 672-002 - Statistical Inference

CSI 672-002 - Statistical Inference
Dr. Kent Miller
Tuesdays 7:20-10 p.m. - Planetary Hall, Room 131

This semester the Department is offering CSI 672-002 - Statistical Inference. This course is be taught face to face by Dr. Kent Miller.   

Fundamental principles of estimation and hypothesis testing. Topics include limiting distributions and stochastic convergence, sufficient statistics, exponential families, statistical decision theory and optimality for point estimation, Bayesian methods, maximum likelihood, asymptotic results, interval estimation, optimal tests of statistical hypotheses, and likelihood ratio tests. Offered by Computational & Data Sciences. May not be repeated for credit.

Statistical inference is a critical skill to have in a data driven world. It is rare to have the financial and human
resources to examine an entire population (e.g. U.S. Census). More common it is to draw a sample from a
larger population (e.g. clinical trial, market survey, opinion poll); and then, based on the sample, to make inferences about the population (e.g. How eff ective is our new medication? How many people will buy our new product? Where do voters stand on a particular issue?). One looks for an underlying probability distribution that best explains the sample, and then one uses that distribution to make estimates and to test hypotheses about the
population.

The instruction will be handled in a lecture/lab format. Statistical Inference is more than a traditional knowledge
transfer course. It also develops hands-on skills with computational tools to help the student solve problems. A
student with modest software skills should be able to complete the course on-line, which may be necessary if
university authorities so order.

Equivalent to STAT 652.

Recommended Corequisite: STAT 554.

Registration Restrictions: Required Prerequisite: STAT 544B-. B- Requires minimum grade of B.