Department of Mathematical Sciences

*Key to listing:* Course code is followed by course name and
credits. In parentheses are shown the number of weekly lecture and
laboratory hours respectively. Finally the semester for which the course
is offered is shown. Unless otherwise indicated, these courses are
offered each year.

**STAT 602 - Experimental Design**
3 Credit (3 + 0) *Alternate Spring*

Constructing and analyzing designs for experimental investigations; completely
randomized, randomized complete
block and Latin-square designs, split-plot designs, incomplete block designs,
confounded factorial designs, nested
designs, treatment of missing data, comparison of designs. (Prerequisites:
STAT 401 or permission of instructor).
Example text:

**STAT 605 - Spatial Statistics**
3 Credit (3 + 0) *As Demand Warrants*

Stochastic processes. Geostatistics including kriging and spatial design of
experiments. Point processes including
model selection and K-functions. Lattice process models and image analysis.
Computer intensive statistical
methods. (Prerequisites: STAT 401 and MATH 202 or permission of instructor.)
Example text: Statistics for
Spatial Data by Cressie.

**STAT 611 - Time Series**
3 Credit (3 + 0) *As Demand Warrants*

An applied course in time series and repeated measure analysis. Autoregression and
moving average models.
Estimation of parameters and tests. Prediction. Spectral analysis. (Prerequisites:
STAT 401 or permission of
instructor.) Example text: Time Series by Kendall and Ord.

**STAT 621 - Distribution-Free Statistics**
3 Credit (3 + 0) *As Demand Warrants*

Methods for distribution-free (non-parametric) statistical estimation and testing.
These methods apply to many
practical situations including small samples and non-Gaussian error structures.
Univariate, bivariate, and
multivariate tests will be presented and illustrated using a variety of applications
and data sets. (Prerequisites:
STAT 200 (Juneau STAT 373); STAT 401 recommended, or permission of instructor).
Example text: Practical
Nonparametric Statistics by Conover.

**STAT 631 - Categorical Data Analysis**
3 Credit (3 + 0) *Alternate Fall*

Statistical methods designed for count and categorical data. Contingency tables.
Logistic and related models.
Loglinear models. Repeated categorical responses. Survival data. (Prerequisites:
STAT 401 or permission of
instructor). Example text: Categorical Data Analysis by Agresti.

**STAT 640 - Exploratory Data Analysis**
3 Credit (3 + 0) *As Demand Warrants*

Quantitative and graphical methods for explaining data and for presenting
data to others. Computer-aided
detection and analysis of patterns in data. Methods for analysis of patterns
in data. methods for validating the
assumptions of common statistical tests and models. Use of computer graphics
in statistical analysis.
(Prerequisite: STAT 200 (Juneau STAT 373); STAT 401 recommended).
Example text: Typically a text by
Tukey.

**STAT 651 - Statistical Theory I**
3 Credit (3 + 0) *Fall*

Probability, distributions of random variables,
conditional probability and stochastic independence, distributions
of functions of random variables, expectation,
limiting distributions, moment generating functions, distributions
derived from the normal distributions.
(Prerequisites: MATH 202, MATH 314, STAT 200, 300, or MATH 371,
STAT 401 recommended.) Example text:
Mathematical Statistics and Data Analysis by Rice.

**STAT 652 - Statistical Theory II**
3 Credit (3 + 0) *Spring*

Estimation of parameters including evaluation of efficiency and sufficiency,
maximum likelihood and method of
moments estimation, bootstrap and other resampling techniques to estimate
variances, and construction of
confidence intervals. Hypothesis testing including the Neyman-Pearson
paradigm and likelihood ratio tests for
evaluating one and two sample problems, the analysis of categorical data,
and analysis of variance problems.
Frequentist and Bayesian inference. (Prerequisites: STAT 651.) Example
text: Mathematical Statistics and Data
Analysis by Rice.

**STAT 653 - Statistical Theory III**
3 Credit (3 + 0) *Fall*

Best linear unbiased estimation, Gauss-Markov theory and applications,
maximum likelihood estimation for linear
models, multivariate normal distributions, linear regression and
analysis of variance, weighted regression, robust
and nonlinear regression, logistic regression, Poisson regression,
ridge regression, smoothing, simple time domain
models. (Prerequisites: STAT 652 or MATH 408 and STAT 401 and MATH 314.)
Example text: An
Introduction to Computational Statistics: Regression Analysis by Jennrich.

**STAT 654 - Consulting Seminar**
3 Credit (3 + 0) *Spring*

Topics related to recent consulting problems. Students will be involved
in consulting for other graduate students
and faculty and in learning about consulting practices. (Prerequisite:
admission into the interdisciplinary
graduate program in statistics). May be repeated for credit up to 3 credits.
Example text: none.

**STAT 661 - Sampling Theory**
3 Credit (3 + 0) *Alternate Spring*

Statistical theory for sampling and sample surveys. Choice of method, power, and sample
size considerations,
treatment of sampling and non-sampling biases. Sampling methods based on detectability.
Adaptive sampling.
Spatial sampling. Mark and recapture methods. The jackknife, the bootstrap, and
resampling plans.
(Prerequisites: STAT 200 (Juneau STAT 373); STAT 401 recommended, or permission
of instructor). Example
text: Sampling by Thompson.

**STAT 680 - Data Analysis in Biology**
3 Credit (2 + 3) *As Demand Warrants*

Biological applications of nonparametric statistics, including tests
based on binomial and Poisson distributions,
analysis of two-way and multiway contingency
tables, and tests based on ranks; multivariate
statistics, including
principle component analysis ordination techniques,
cluster analysis, and discriminate analysis; and time-series
analyses. Introduction to the use of the computer,
and use of statistical packages. Each student will analyze a
data set appropriate to the student's research interests.
(Prerequisites: STAT 300, 401 and either graduate
standing in a biologically oriented field or permission of instructor). Example text: none.