Registry

Module Specifications

Archived Version 2009 - 2010

Module Title	Time Series
Module Code	MS447
School	School of Mathematical Sciences
Online Module Resources
Module Co-ordinator	Professor John Appleby	Office Number	X133
Level	4	Credit Rating	7.5
Pre-requisite	None
Co-requisite	None

Module Aims

The module aims to explain fundamental concepts of a generalised linear model (GLM) and explain the situations in which such a model will apply. It also seeks to introduce the main conceptsunderlying the analysis of Time Series models. It aims to cover the syllabus of the Time Series part of the Institute and Faculty of Actuaries Core Technical subject CT6.

Learning Outcomes

On completion of this model students should be able to:

Understand the role and scope of the General Linear Model;
Apply time series models to problems in insurance and finance;
Diagnose the type of a time series from data;
Be able to apply Monte-Carlo simulation techniques to problems in non-life insurance mathematics.

Indicative Time Allowances
	Hours
Lectures	36
Tutorials	12
Laboratories
Seminars
Independent Learning Time	64.5

Total	112.5
Placements
Assignments

NOTE

Assume that a 7.5 credit module load represents approximately 112.5 hours' work, which includes all teaching, in-course assignments, laboratory work or other specialised training and an estimated private learning time associated with the module.

Indicative Syllabus

Generalised Linear Model: Exponential family of distributions, link function. Prediction. Parameter estimation in a GLM. Deviance and deviance residuals. CT6 - [viii]

Time Series: Stationary, integrated time series. Filtering. Linear time series: AR, ARMA, ARIMA. Random walk theory applied to time series. CT6 - [ix] 1 -11

Applications of Time Series: Diagnostics of time series. Applications of theory to investment variables and financial time series. Deterministic forecasts. CT6 - [ix] 12 -13

Monte Carlo simulation: Pseudo-random number generation. Generation of random variates. Variance reduction. Simulation of time series. Reliability of simulation. CT6 - [x]