Division of Statistics
Dalhousie University
Department of Mathematics and Statistics
Dalhousie University, Halifax, Nova Scotia, Canada
Tel: (902) 494 - 2572
Fax: (902) 494 - 5130
[ Home ]     [ Research ]     [ People ]     [ Programs ]    [ Resources ]

# Statistics Director Dalhousie University Halifax, N.S., Canada B3H 3J5 Phone: (902)494-2572 FAX: (902)494-5130 E-mail: statdir@mathstat.dal.ca Additional information is available on the Registrar's Office Statistics Page Registration information, university calendar and timetable are available at the Registrar's Office Home Page

Courses:

STAT 1060.03 : Introductory Statistics for Science and Health Sciences. This class gives an introduction to the basic concepts of statistics through extensive use of real-life examples drawn from a variety of disciplines. The first part of the class is about designing experiments properly and then describing and summarizing the results of the studies by using descriptive statistics. From there we move to analyzing relationships between variables. In the final part of the class, we develop the basics of statistical inference explaining how to make valid generalizations from samples to populations. Both estimation and hypothesis testing are carried out for one and two sample problems for both means and proportions as well as for simple linear regression. Students will learn to use the statistical package MINITAB. The natural sequel for this class is STAT 2080.03. Other possibilities are STAT 2060.03 and STAT 2050.03. Credit will not be given for STAT 1060.03 if credit has previously been obtained for STAT 2060.03.

STAT 2060.03 : Introduction to Probability and Statistics. Rigorous introduction to probability and statistical theory. Subject matter is developed systematically beginning with the fundamentals of probability and following with statistical estimation and testing. The interrelationship between probability theory, mathematical statistics and data analysis will be emphasized. Topics covered include elementary probability, random variables, distributions, estimation and hypothesis testing. Estimation and testing are introduced using maximum likelihood and the generalized likelihood ratio. Natural sequels for this class are STAT 2080.03 and 3360.03

STAT 2080.03 : Statistical Methods for Data Analysis and Inference. The usual sequel to STAT 1060.03 or STAT 2060.03. This class introduces a number of techniques for data analysis and inference commonly used in the experimental sciences. The class begins with an introduction to model building in linear models and develops the techniques required for multiple regression. From here we consider analysis of variance, factorial designs, analysis of covariance using the general techniques for linear models. The last part of the class will include techniques for two and three way tables along with logistic regression. The use of a computer package for carrying out the computations will be an integral part of the class. Students will design and carry out a simple experiment as part of this class. A natural sequel for this class is STAT 3340.03 or STAT 3345.03.

STAT 2050.03 : Exploratory Data Analysis. This class is designed to introduce the student to exploratory data analysis and graphical techniques making extensive use of statistical software such as S-plus. Data sets from both experimental and observational studies will be used extensively and the emphasis will be on finding patterns and structure in the data. The student completing the class will be able to do sophisticated graphing, data reduction and data handling. The skills learned will be very useful in several of the advanced statistics classes.

STAT 3340.03 : Regression and Analysis of Variance. A thorough treatment of the theory and practice of regression analysis. Topics include: fitting general linear models using matrices, optimality of least squares estimators (Gauss-Markov theorem), inferences, simple and partial correlation, analysis of residuals, case-deletion diagnostics, polynomial regression, transformations, use of indicator variables for analysis of variance and covariance problems, model selection, and an introduction to nonlinear least squares. This class makes extensive use of computer packages.

STAT 3345.03 : Environmental Risk Assessment. Statistical methods for assessing risk are discussed, including dose-response models, survival analysis, relative risk analysis, bioassay, estimating methods for zero risk trend analysis and association risks. Case studies are used to illustrate the methods.

STAT 3350.03 : Design of Experiments. The aim of the class is to develop the fundamental statistical concepts required for designing efficient experiments to answer real questions. The first main subject is unit variation and control. The basic concepts of replication, blocking and randomization are each examined. The second main subject is treatment questions and structure. The ideas of factorial designs, split-plot and incomplete plot designs are presented. We conclude with a look at response surface methodology.

STAT 3360.03 : Probability. The concepts and application of probability. Topics include the classical discrete and continuous distributions, including the binomial, hypergeometric, multinomial, Poisson, uniform, exponential and normal; definitions and properties of random variables; independence; sums of independent random variables, including the law of large numbers and central limit theorem; conditional probability; and the bivariate normal distribution. Examples will be taken from the natural and physical sciences.

STAT 3380.03 : Sample Survey Methods. The development of design and analysis techniques for sample surveys. Topics include simple, stratified and systematic random sampling, ratio and regression estimation, sub-sampling with units of equal and unequal size, double-multistage and multiphase sampling, non-sample errors and non-respondents.

STAT 3460.03 : Intermediate Statistical Theory. This class provides an intermediate level coverage of statistical theory to provide a framework for valid inferences from sample data. The methods developed are based on the likelihood function and are discussed from the frequentist, likelihood, and Bayesian approaches

>> Degree
Programs

>> Contact Info

>> Courses