• Title: Forecasting with Generalised Additive Models (GAMs) in R

• Date: 21 February 2024

• Time: 14:00 - 16:00 (GMT)

• Instructor: Dr Nicola Rennie (Lancaster University)

• Slides: nrennie.github.io/f4sg-gams

• GitHub: github.com/nrennie/f4sg-gams

• Posit Cloud: posit.cloud/content/7637971

• Anyone who is interested in extending their knowledge beyond simple regression models and learning more about generalised additive models.
• Newcomers to the field of generalised additive models who want to understand their importance and relevance in forecasting.
• Academics, students, data scientists, researchers, and practitioners in the field who are working with data containing complex nonlinear relationships.
• People who want to learn how to implement generalised additive models using R.

By the end of the workshop, participants will:

• know what generalised additive models are;
• understand why and when they might be appropriate for certain types of data;
• be able to fit and evaluate GAMs using the {mgcv} package in R;
• understand how to interpret the output from fitted models.

• Basic knowledge of R.
• Basic knowledge of statistics and linear modelling.

• This session will provide an overview of generalised additive models (GAMs), demonstrate the practical aspects of fitting such models, and describe how to evaluate and interpret different them.
• Live demonstrations and hands-on coding exercises will give participants the opportunity to practice implementing models using R.

• Introduction to the data and the {mgcv} package (15 minutes)
• Fitting and evaluate GAMs using the {mgcv} package (15 minutes)
• Forecasting using GAMs (15 minutes)

Required:

• {mgcv}
• {gratia}

Optional:

• {COVID19}
• {tidyverse}