Practical use of the software package R. Multiple regression. Analysis of variance. Analysis of categorical data. Generalised linear models. Basic principles for statistical inference. Simulation from a model. Properties of estimators. Statistical power. Numerical maximisation of the likelihood function. Some asymptotic results. Model selection. Non-parametric tests.

## Learning outcomes

1. Knowledge.

The student has an overview over the underlying assumptions and practical applications of multiple regression, analysis of variance, analysis of categorical data and generalised linear models using the software package R. In addition, the student knows how simulation methods can be used in finding properties of estimators, in hypothesis testing and in computation of power. The student also has basic knowledge about numerical methods for fitting non-standard models using maximum likelihood and how results from asymptotic theory can be used in estimating uncertainty in parameter estimates and in hypothesis testing.

2. Skills.

The student can handle and analyse collected datasets using the software package R. The student is also capable of formulating simple, non-standard statistical models, implementing the model in computer code and fitting such models using standard numerical optimisation algorithms. In both situations, the student is able to assess the properties of a given method using simulations or using methods based on asymptotic theory.

## Prerequisites

##### Recommended previous knowledge

ST0103 Statistics with applications, MA0001 Mathematical Methods A and MA0002 Mathematical methods B or equivalent.

## Files/Documents

## ISCED Categories