Availability:

• This is a core module for the MSc in Mathematical Finance.
• Not available to undergraduate students.
• PhD students interested in taking the module should consult the lecturer.

Commitment:

• 30 hours of lectures and 8 hours of lab sessions

Content:

• Part 1: Classical and Bayesian methods of statistical inference (weeks 1-5)
  • Properties of random samples
  • Statistics, sufficiency and likelihood
  • Point estimation, maximum likelihood estimation
  • Hypothesis testing and interval estimation
  • Elements of Bayesian inference
  • Linear models
• Part 2: Time Series (weeks 6-10)
  • Auto-regressive and moving average models (ARMA), unit root (ARIMA) and seasonal models (S-ARIMA), heteroscedastic models (GARCH and extensions such as EGARCH, GARCH-M,...) and an introduction to stochastic volatility models.
  • Linear and non-linear modelling of financial time series with R: exploratory analysis, model selection, model fitting, model validation and forecasting.
  • Illustrative financial applications.

Assessment:

• 1 x 2 hour exam at 80%
• 1 x 15min class test on part 1 of course at 5%
• 1 x project in R on part 2 of course at 15%

Illustrative Bibliography: