Introduction to modern Generalised Additive Models in R

Date: Friday 24 September 2021, 9.30AM - 4.30PM
Location: London
CPD: 6.0 hours
RSS Training
Book now


Share this event

Level: Professional (P)


Generalized Additive Models (GAMs) models are an extension of traditional regression models, and have proved to be highly useful for both predictive and inferential purposes in a variety of scientific and commercial applications. One reason behind the popularity of GAMs is that they strike a balance between flexibility and interpretability, while being able to handle large data sets. The thought part of the course will provide an overview of GAM theory, methods and software, while the hands-on sessions will make sure that the attendees will be ready to start doing GAM modelling in R as soon as the course is over. 

Please note: Bookings will close 4 working days before the course start date or when the course has reached its maximum capacity.
 

Level: Professional (P)


Generalized Additive Models (GAMs) models are an extension of traditional regression models, and have proved to be highly useful for both predictive and inferential purposes in a variety of scientific and commercial applications. One reason behind the popularity of GAMs is that they strike a balance between flexibility and interpretability, while being able to handle large data sets. The thought part of the course will provide an overview of GAM theory, methods and software, while the hands-on sessions will make sure that the attendees will be ready to start doing GAM modelling in R as soon as the course is over.


Learning Outcomes

Attendees with no experience in GAM modelling will get an understanding of what GAM models are, when are they useful and how can they be used to perform statistical analysis, for inferential or predictive purposes. Attendees who have some experience with GAMs will learn about the new Big Data and visual GAM methods, as well as about GAMLSS models and quantile GAMs.

Topics Covered

  • model building, inference and fitting methods 
  • key Bayes empirical smoothing theory
  • types of smooth and random effects
  • visual methods and diagnostics 
  • Generalized Additive Models for Location Scale and Shape
  • Quantile GAMs
  • GAM modelling in R


Target Audience

The target audience are practictioners, either in industry or academia, interested in learning new powerful statistical methods, which can be used in a wide variety of applications such as rainfall modelling, electricity demand forecasting, survival analysis and air pollution modelling to name a few.


Knowledge Assumed

Attendees should have some background on (linear) regression modelling, and a good understanding of fundamental statistical concepts such as probability densities, quantiles, etc. Some basic proficiency with R (eg. loading data, accessing data frames, basic use of the lm() function), at a level equivalent to a couple of days of self-study, is also assumed. 

Attendees will need to bring a laptop with a recent version of R installed. Prior to the course attendees will be asked to install some additional R packages.
 

Dr. Matteo Fasiolo

Dr. Matteo Fasiolo is an EPSRC Doctoral Prize fellow at the School of Mathematics at the University of Bristol.His current research is concerned with extending Generalized Additive Models (GAMs), with particular focus on electricity load forecasting applications. He is the authors of the qgam R package, which provides fitting methods for quantile regression GAMs, and of the mgcViz R package, which offers new visualization tools for GAMs. He is also the author of the mvnfast, synlik and esaddle R packages on CRAN. Matteo studied an industrial and financial engineering, before obtaining a PhD in Statistics at the University of Bath.

 

Prof. Simon Wood

Prof. Simon N. Wood is Professor of Statistical Science at the University of Bristol. He mainly research focus are modern regression modelling, especially using smooth functions and random effects, and statistics applied to ecology, especially to ecological dynamics. He is the author of the recommended R package mgcv and of the corresponding book “Generalized Additive Models: an introduction with R”. Currently he is particularly interested in spatio-temporal modelling, sparse methods, scalable statistical computing for big models and data, and model selection issues. Recent applications have been in air-pollution modelling and electricity demand prediction.

 

Fees

   

Registration before 
24 August 2021

 

Registration on/after
24 August 2021

                                  


Non Member 

RSS Fellow 

RSS CStat/GradStat/Data Analyst 
also MIS & FIS

 

£392.00+vat 

£332.00+vat 

£314.00+vat

£435.00+vat 

£370.00+vat 

£349.00+vat

 
Book now