Statistical Inference

Overview

Statistical investigations. Initial data analysis. Sample diagnostics. Point estimation: maximum likelihood, least squares. Multiple linear regression. Significance tests: Neyman-Pearson approach, likelihood ratio tests. Confidence intervals. Introduction to Experimental design. Bayesian methods. Oral presentation of aspects of statistics.

Learning Objectives

On completion of the module, it is intended that students will be able to: understand how to build statistical models, know the issues involved with using real data and use sample diagnostic methods to test data for independence, normality and goodness of fit; understand the difference between estimates and estimators in terms of finding an unknown parameter and understand the assessment of an estimator's unbiasedness, relative efficiency, mean square error, sufficiency and whether a distribution belongs to the regular exponential class; understand and use the method of moments, of maximum likelihood and of least squares to provide unbiased estimates of distribution parameters; understand and use experimental design, the different forms of analysis of variance techniques and use and interpret methods of association; rates, relative risk and the odds ratio; through expanding their knowledge of significance and hypothesis tests, formulate confidence intervals and regions and use pivotal quantities; understand Bayesian inference; the prior and posterior distributions, and degree of belief; be able to carry out a statistical investigation of real data in group work using SAS and present the results in a written and oral presentation.

Skills

Statistical modelling and problem solving. Application of statistical methods in data analysis. Presentation skills.

Assessment

SAS coursework; presentation.