Week 5

Overview #

We will continue with the Time Series component of the subject, which corresponds to the Section 2 of the CS2 syllabus; In particular, for Module 9 we will

• study in detail \(ARMA(p,q)\) models;
• study in detail \(ARIMA(p,d,q)\) models;
• introduce seasonal \(ARIMA\) models;
• introduce multivariate autoregressive \(VAR(p)\) models;
• introduce some additional models mentioned in the CS2 syllabus.

In terms of Module 10 we will

• discuss how to fit the models seen in Module 9 to data sets;
• discuss how to use those fitted models to perform forecasts.

Detailed learning outcomes for Module 9 correspond to 2.1.1, 2.1.5, 2.1.7, 2.1.8, and 2.2.2 of the CS2 syllabus here.

Detailed learning outcomes for Module 10 correspond to 2.2.1 and 2.2.4 of the CS2 syllabus here.

Main references and lectures #

Module 9 #

Read: Module 9 SLIDES

annotated slides (UG) | annotated slides (PG)

Watch: refer to your lecture recording under “Lecture Capture” ( UG/ PG). This is where annotated slides will be made available, too.

Module 10 #

Read: Module 10 SLIDES

annotated slides (UG) | annotated slides (PG)

Watch: refer to your lecture recording under “Lecture Capture” ( UG/ PG). This is where annotated slides will be made available, too.

Additional preparation and resources #

Mandatory #

1. Module 9: Chapter 3.1, 3.3, 3.6, 3.9, 5.6 of Shumway and Stoffer (2017)
2. Module 10: Chapter 3.4, 3.5, and 3.7 of Shumway and Stoffer (2017)
3. S6 (CS2 Time Series notes)

Optional #

Tutorials #

Note that a full list of exercises is available here. Please let us know of any mistake or required update on Ed.

Pre-Tutorial work #

Please study those questions before the tutorial.

Pre-Tutorial exercises are available in the Pre-Tutorial book, which already includes solutions. It is recommended to attempt the questions before looking at the solutions

Tutorial materials #

Some questions have been especially selected for the tutorials. Students should review and attempt those questions prior to their scheduled tutorial, after they complete the pre-tutorial work.

The Tutorial book includes all questions for the whole semester already, but solutions will only be added sequentially at the end of each week, as we work our way through the set.

Note that solutions will be gradually added to that same document. Hence it is not recommended to print it, as it will regularly change (typos will also dynamically be corrected).

This week #

This week, we will finish off with questions of Module 9 and 10.

Next week (week 6) #

Next week, we will discuss start the claims modelling component of the subject with Module 2.

Additional questions #

The “additional questions” are here for reinforcement or revision, but are not the main focus of the tutorials. Solutions for those exercises are already available.

Preparation for assessment #

Mid-semester (15%) and final (60%) exams #

• Finalise your summaries of Modules 7 and 8
• Your summary of Module 9 should be quite advanced; start the summary of Module 10.

Note that all of Modules 7 to 10 are in scope for the mid-semester exam; that includes tutorials of week 5.

You could start preparing for the mid-semester exam by consolidating your summaries over the Easter break, and look at past CS2 exams (look for questions classified as “TS”); refer to the mappings available on the website here.

Assignment (25%) #

• The assignment is now available. You should aim to finalise your data analysis by the end of the Easter Holidays at the latest. Allow at least one week for preparing and recording your video.

References #