MATH5824 Generalised Linear and Additive Models

Author

Robert G Aykroyd

Published

May 6, 2024

Weekly schedule

Items will be added here week-by-week and so keep checking when you need up-to-date information on what you should be doing. Note that items specific to MATH5824M will be marked accordingly, otherwise items refer to the material common with MATH3823.

Week 11 (6 - 10 May)

Revision and examination preparation. Please do not forget that the examination this year is closed-book and all questions and you will be expected to attempt all questions.

Please note that the MATH5824 Lecture on Friday 10 May is CANCELLED.

Week 10 (29 April - 3 May)
  • Before next Lecture: Prepare for Exercises.
  • Lecture on Tuesday: Selected Exercises from Chapter 6.
  • Lecture on Thursday: Selected Exercises from Chapter 5.
  • Lecture on Friday: Revision and examination preparation.
Week 9 (22 - 26 April)
  • Submit your Computer Practical report on-time!
  • Before next Lecture: Re-read Chapter 6: Log-linear Models.
  • Lecture on Tuesday: Start Chapter 7: Extensions to Loglinear models.
  • Lecture on Thursday: Continue Chapter 7: Extensions to Loglinear models.
  • Lecture on Friday: Cover Chapter 6: General Additive Models.
  • Weekly feedback: Check any previous Exercises and solutions not completed.
Advanced notice

Please note that I will not be available after Friday 22 March until Monday 22 April.

Assessed Practical
  • Module Assessment: Set on 12 March with submission deadline 23 April (that is after the break). You will be expected to write a short report based on an RStudio practical.
  • Computer classes: Supervised session on 19/20 March – check your timetable.
  • Generative AI usage within this module: The assessments for this module fall in the red category for using Generative AI which means you must not use Generative AI tools. The purpose and format of the assessments makes it inappropriate or impractical for AI tools to be used.
Week 8 (18 - 22 March)
  • Before next Lecture: Re-read Chapters 4 and 5 ready for practical.
  • Lecture on Tuesday: Start Chapter 6: Loglinear Modelling with Sections 6.1-6.3.
  • Computer Practical on Tuesday/Wednesday: Work on Assessed Practical.
  • Lecture on Thursday: Complete Chapter 6: Loglinear Modelling with Section 6.4-6.5.
  • Lecture on Friday: Complete Chapter 5: Choosing the smoothing parameter with Sections: 5.4-5.6.
  • Weekly feedback: Start Exercises for Chapter 6 and check answers with solutions.
Week 7 (11 - 15 March)
  • Before next Lecture: Re-read Chapter 5: Sections 5.1-5.3.
  • Lecture on Tuesday: Complete Chapter 5: by looking at Section 5.4: Odds Ratio and Section 5.5 - Application to dose-response experiments.
  • Lecture on Thursday: Consider selected Exercises from previous Chapters and discuss Assessed Practical. If time start Chapter 6: Log-linear Modelling.
  • Before next Lecture: Please re-read MATH5824 Chapter 4: Smoothing splines.
  • Lecture on Friday: Start Chapter 5: Choosing the smoothing parameter with Sections: 5.1-5.3.
  • Weekly feedback: Complete Exercises for Chapter 5.
Week 6 (4 - 8 March)
  • Before next Lecture: Re-read Chapter 4: Sections 4.1-4.2.
  • Lecture on Tuesday: Complete Chapter 4 with
    Sections: 4.3 Model deviance, 4.4 Model Residuals & 4.5 Fitting GLMs in R
  • Lecture on Thursday: Start Chapter 5: Modelling Proportions with Sections 5.1 & 5.2.
  • Before next Lecture: Please re-read MATH5824 Section 4.1: Overview and Section 4.2: The penalized least-squares criterion.
  • Lecture on Friday: Complete Chapter 4 with Sections 4.3-4.5.
  • Weekly feedback: Complete Chapter 4 Exercises.
Week 5 (26 February - 1 March)
  • Before next Lecture: Re-read all of Chapter 3.
  • Computer Practical: Either on Tuesday or Wednesday, attend supervised computer class – see your timetable. For information see Week 5 Computer Practical folder on Minerva.
  • Lecture on Tuesday: Start Chapter 4 by covering Section 4.1: The identically distributed case.
  • Lecture on Thursday: Chapter 4: Section 4.2: The general case.
  • Lecture on Friday: Complete Chapter 4 with Section 4.1: Overview, then cover Section 3.4: Roughness penalties which was forgotten in the previous lecture. Then, continue with Section 4.2: The penalized least-squares criterion.
  • Weekly feedback: Check your answers on Chapter 3 Exercises with online solutions and start MATH5824M Chapter 3 Exercises.
Week 4 (19 - 23 February)
  • Before next Lecture: Be confident with all material up to, and including, Section 3.2: The GLM structure.
  • Lecture on Tuesday: We will cover Section 3.3: The random part of a GLM, Section 3.4: Moments of exponential-family distributions and, if time permits, Sections 3.5: The systematic part of the model.
  • Lecture on Thursday: Complete Chapter 3 by covering Section 3.6: The link function. Then, we will consider selected Exercises from Section 3.7.
  • Before next Lecture: Please re-read the whole of MATH5824M Chapter 2: Introducing Splines.
  • Lecture on Friday: We will start MATH5824M Chapter 3: Interpolating Splines by looking at Sections 3.1-3.3, Overview, Natural splines and Properties of natural splines. [Edit: Also, Section 3.4: Fitting interpolating splines in R.]
  • Weekly feedback: Complete the first Chapter 3 Quiz and start the MATH5924M Exercises in Section 3.6 and the MATH3823 Exercises in Section 3.7.
Week 3 (12 - 16 February)
  • Before next Lecture: Please re-read Section 2.5: Model shorthand notation and Section 2.6 Fitting linear models in R, and read Section 2.7: Ethics in statistics and data science.
  • Lecture on Tuesday: We will start Chapter 3 with Section 3.1: Motivating examples and Section 3.2: The GLM structure.
  • Before next Lecture: Please re-read Sections 3.1 and 3.2 carefully.
  • Lecture on Thursday: CANCELLED due to illness.
  • Lecture on Friday: CANCELLED due to illness.
  • Weekly feedback: Complete the *MATH5824M Exercises** in Chapters 1 & 2.
Week 2 (5 - 9 February)
  • Before next Lecture: Please re-read Section 2.1: Overview and Section 2.2: Linear models, and self-study Section 2.3: Types of normal linear model.
  • Lecture on Tuesday: We will cover Section 2.4: Matrix representation of linear models and briefly Section 2.5: Model shorthand notation.
  • Before next Lecture: Please re-read Sections 2.4 and 2.5 carefully.
  • Lecture on Thursday: We will cover Section 2.6: Fitting linear models in R then discuss selected Exercises from Chapters 1 and 2.
  • Before next Lecture: Please re-read the whole of MATH5824M Chapter 1: Non-parametric Modelling.
  • Lecture on Friday: We will cover the whole of MATH5824M Chapter 2: Introducing Splines.
  • Weekly feedback: Complete the Chapter 2 Quizzes and complete the Exercises in Section 2.8. Also, complete the MATH5824M Section: 1.4 Exercises.
Week 1 (29 January - 2 February)
  • Before first Lecture: Please read the Overview.
  • Lecture on Tuesday: We will briefly cover all material in Chapter: Introduction.
  • Before next Lecture: Please re-read Chapter 1 carefully, especially any sections not covered in Lectures.
  • Lecture on Thursday: Start Chapter 2: Essentials of Normal Linear Models with Section 2.1: Overview & Section 2.2: Linear models.
  • Before next Lecture: Please read the MATH5824M Overview.
  • Lecture on Friday: We will cover the whole of MATH5824M Chapter 1: Non-parametric Modelling.
  • Weekly feedback: Complete the Chapter 1 Quizzes and self-study the Exercises in Section 1.5 – solutions to be added during Week 1. If you have time, then also self-study the MATH5824M Exercises in Section 1.4.
Advanced notice
  • Module Assessment: Set on 14 March with submission deadline 23 April (that is after the break). You will be expected to write a short report based on an RStudio practical.
  • Computer classes: 27/28 February for Practice and 19/20 March for Assessment – check your timetable.
  • Generative AI usage within this module: The assessments for this module fall in the red category for using Generative AI which means you must not use Generative AI tools. The purpose and format of the assessments makes it inappropriate or impractical for AI tools to be used.
Provisional Weekly Lecture Schedule
Week 1 Chapter 1 All
Week 2 Chapter 2 All
Week 3 Chapter 3 Sections 3.1-3.3
Week 4 Sections 3.4-3.5
Week 5 Chapter 4 Sections 4.1-4.3
Week 6 Sections 4.4-4.5
Week 7 Chapter 5 Sections 5.1-5.3
Week 8 Sections 5.4-5.6
Easter
Week 9 Chapter 6 All
Week 10 Exercises All Exercises
Week 11 Revision