Скачать с ютуб Introduction to Generalized Additive Models with R and mgcv в хорошем качестве

Introduction to Generalized Additive Models with R and mgcv

Scientists are increasingly faced with complex, high dimensional data, and require flexible statistical models that can accommodate them. There is often a tension between applying simple models that don't fit the data that well but are really easy to interpret, and using more black-box machine learning methods that do fit the data well but are hard to interpret. Generalized Additive Models (GAMs) fit into the gap between these two extremes, using highly interpretable splines to model non-linear relationships between covariates and response that are learned from the data. The mgcv package for R is one of the most popular packages for fitting smooth, non-linear relationships, providing a wide range of powerful tools for modelling complex data. However, many scientists are not familiar with GAMs, how they learn from data to fit non-linear relationships, nor how to use the mgcv software to fit the models in interpret the results. In this online webinar I will introduce participants to splines and how GAMs use splines to learn from the underlying data. I'll show you how splines work and describe the different types of splines available in mgcv and what they can be used for. In addition, I'll cover * model fitting in R with mgcv, * model checking and diagnostics, * plotting and working with GAMs. Plus I'll give examples of the range of statistical models and data types that can be handled and modelled within mgcv. Participants will be assumed to be familiar with the basics of R (such as loading and manipulating data, and plotting) and regression in R (lm() and glm()). The webinar is free to attend; this is a difficult time for everyone, monies are tight, and options for training have been reduced because of Covid-19. However, if you are financially able, please consider making a donation to the University of Regina Student Emergency Fund, which helps provide urgently needed support to our hardworking students. You can donate to the U of R Student Emergency Fund at https://giving.uregina.ca/student-eme....