Lesson 3

library(dplyr)
library(magrittr)
library(ggplot2)
library(tidyr)
theme_set(theme_minimal())

First, let’s slow down a bit

Try to complete Task 3 and 4 from Lesson 2 (and possibly others from the previous lessons)

Pick projects!

Note on models

• Reuse methodology slides: https://docs.google.com/presentation/d/1jkcKfTOupedZRoJZ9qSuGVDBjlzZFt6vOyEyN0pwduY/edit?usp=sharing

Linear regression

• The most ubiquitous type of model — see e.g. https://lindeloev.github.io/tests-as-linear/
• Taylor series
• Intercept, coefficients
• Dummy coding
  • There are other ways to code (e.g. effect coding)
  • model.matrix in R does the coding for you
• Linear predictors and matrix multiplication
  • Intercept in the matrix
• Least squares and its relatio to normal distribution
  • Maximum likelihood estimator is equivalent to least-squares.
• Generalized linear models (GLM)
  • Link function, inverse link function
  • Most common link functions:
    • log for positive outcomes (i.e. exponentiate the predictors),
    • logit (i.e. apply inverse logit to the predictor) for outcomes in [0, 1] - most notably for probabilities in logistic regression.
• Posterior predictive checks as a method to determine which predictors to add

Now, let’s do some tasks. Note that they match workflow (simple stuff first, …)