Module Specifications..
Current Academic Year 2023  2024
Please note that this information is subject to change.
 
Repeat examination 

Description MS228 aims to provide a strong foundation in the fundamental statistical method of regression modelling. Simple and multiple linear regression models will be presented. The fitting and interpretation of regression models will be explained and practical examples given. The linear model will be extended to model non normal data using generalised linear models (GLMs). The regression models will be applied to practical datasets using R. Students will also be introduced to Bayesian statistical methods and their use in credibility theory.  
Learning Outcomes 1. Perform exploratory data analysis techniques on a sample of data. 2. Describe and interpret linear regression models and estimate the parameters of a simple linear regression model. 3. Describe and interpret generalised linear models (GLMs). 4. Fit linear and generalised linear models to datasets using R and apply appropriate model checks. 5. Understand the Bayesian approach to statistical modelling. 6. Apply the Bayesian and empirical Bayesian approaches to credibility theory.  
All module information is indicative and subject to change. For further information,students are advised to refer to the University's Marks and Standards and Programme Specific Regulations at: http://www.dcu.ie/registry/examinations/index.shtml 

Indicative Content and Learning Activities
Exploratory Data Analysis Data visualisation – histograms, box plots, scatter plots, heatmaps. Use of R visualisation packages. Correlation measures. Introduction to Principal Components Analysis. Use of R to implement analysis techniques. Linear Regression Exploratory data analysis; correlation; least squares estimation; model fitting in R; goodness of fit measures; residual checking; predication; confidence intervals. Multiple Linear Regression Model definition; modelling fitting and checking in R. Generalised Linear Models (GLM) Exponential family  binomial, Poisson, exponential, gamma and normal; link function and canonical link function; deviance and scaled deviance; model fitting in R; Pearson Chisquare test and the Likelihood ratio test; model interpretation. Bayesian Statistics Bayes theorem; prior and posterior distributions; loss functions. Credibility Theory Credibility premium; Bayesian approach to credibility theory; empirical Bayes approach to credibility theory.  
 
Indicative Reading List  
Other Resources None  
Programme or List of Programmes  
Archives: 
