This unit is an introduction to the statistical theory and practice of time series analysis. A time series is a set of data indexed by time. A time series is modelled as a single 'realisation' or sample of a stochastic process, ie a collection of (possibly) dependent random variables. The unit looks at suitable models for time series, and examines the estimation of parameters in these models, hypothesis testing (and alternatively estimating the number of parameters), prediction of future values of the time series (forecasting), models for multivariate time series, and the estimation of periodicity. There will also be a limited look at modelling stochastic volatility. Emphasis in this unit... -- Course Website
Instructor: Professor Barry Quinn
Prerequisites: Admission to MAppStat or PGDipAppStat or PGCertAppStat
