This chapter teaches you how generalized linear models are an extension of other models in your data science toolbox. The chapter also uses Poisson regression to introduce generalize linear models.

Limitations of linear models

Assumptions of linear models

Refresher on fitting linear models

Poisson regression

Fitting a Poisson regression in R

Comparing linear and Poisson regression

Intercepts-comparisons versus means

Basic lm() functions with glm()

Applying summary(), print(), and tidy() to glm

Extracting coefficients from glm()

Predicting with glm()

GLMs, an extension of your regression toolbox

This chapter covers running a logistic regression and examining the model outputs.

Overview of logistic regression

Fitting a logistic regression

Examining & interpreting logistic regression outputs

Bernoulli versus binomial distribution

Bernoulli versus binomial

Simulating binary data

Long-form logistic regression input

Wide-form input logistic regression

Comparing logistic regression outputs

Link functions-Probit compared to logit

Probit versus logit

Fitting probits and logits

Simulating a logit

Simulating a probit

Logistic Regression

This chapter teaches you about interpreting GLM coefficients  and plotting GLMs using ggplot2.

Poisson regression coefficients

Poisson link

lm vs. Poisson coefficients

Plotting Poisson regression

Poisson regression plotting

Understanding output from logistic regression

Understanding odds ratios

Extracting and interpreting odds-ratios

Odds-ratios & confidence intervals in the Tidyverse

ggplot2 and binomial regression

Default trend lines

Methods for trend lines

Comparing probits and logits

Interpreting and visualizing GLMs 

In this chapter, you will learn how to do multiple regression with GLMs in R.

Meervoudige logistische regressie

Een meervoudige logistische regressie fitten

Twee modellen bouwen

Regressieresultaten vergelijken

De volgorde van variabelen vergelijken

Formules in R

Meerdere hellingen

Intercepten

Meerdere intercepts

Aannames van meervoudige logistische regressie

Paradox van Simpson

Niet-lineaire logistische regressie

Conclusie

Multiple regression with GLMs

Bus Commuter dataset

Lineaire regressie is een werkpaard binnen de statistiek, maar kan sommige soorten complexe data niet aan. Een generalized linear model (GLM) breidt lineaire regressie uit met niet-normale verdelingen, waaronder binomiale en teldata. In deze cursus breid je je data science-gereedschapskist uit met GLM’s in R. Je leert binomiale data te modelleren met logistische regressie en teldata met Poisson-regressie. Ook leer je hoe je deze resultaten interpreteert en visualiseert met ggplot2.

Intermediate Regression in R

Leer binomiale data te modelleren met logistische regressie en teldata met Poisson-regressie.

Generalized Linear Models in R

De cursus Generalized Linear Model breidt je regressietoolbox uit met logistische en Poisson-regressie.

Regressieresultaten vergelijken

Generalized Linear Models in R

Praktische interactieve oefening