APTS Statistical Modelling

Ioannis Kosmidis

Professor of Statistics
Department of Statistics, University of Warwick

ioannis.kosmidis@warwick.ac.uk
ikosmidis.com   ikosmidis ikosmidis_

19 April 2024

Course content

• Model selection

• Beyond generalized linear models

• Nonlinear models

Resources

Materials	Link
Preliminary material	ikosmidis.com/files/APTS-SM-Preliminary
Notes	ikosmidis.com/files/APTS-SM-Notes
Slides: Introduction	ikosmidis.com/files/APTS-SM-Slides-intro
Slides: Model selection	ikosmidis.com/files/APTS-SM-Slides-model-selection
Slides: Beyond GLMs	ikosmidis.com/files/APTS-SM-Slides-beyond-glms
Slides: Nonlinear models	ikosmidis.com/files/APTS-SM-Slides-nonlinear-models
Slides: Latent variables	ikosmidis.com/files/APTS-SM-Slides-latent
Lab 1	ikosmidis.com/files/APTS-SM-Notes/lab1.html
Lab 2	ikosmidis.com/files/APTS-SM-Notes/lab2.html

Space and time

	Room	Building
Lectures	A09	ESLC (54)
Labs	C19	Coates (36)
Breakouts	C01, C13	ESLC (54)
Lunch	Foyer	ESLC (54)

Reproducible code

The notes and the slides have chunks of R code, which should be fully reproducible if executed in in the order they appear, in a single R session from

• a standard R installation with
• the contributed packages MuMIn, modelsummary, lme4, ggplot2, SMPracticals

In order to install those packages click on the arrow below and copy and pages the R code

install.packages(c("MuMIn", "modelsummary", "lme4", "ggplot2","SMPracticals"))

Typos and other contributions

Let me know (either during the lectures or by email) if you find any tyops on the slides, or any of the other materials.

I would welcome contributions of code that replicates the analyses in the notes in other languages (e.g. Python, Julia).

Questions

• I will be making time for questions.

• Lift your hand or speak up at any point.