From the course: R Essential Training Part 2: Modeling Data
Unlock the full course today
Join today to access over 22,400 courses taught by industry experts or purchase this course individually.
Predicting outcomes with Poisson or log-linear regression
From the course: R Essential Training Part 2: Modeling Data
Predicting outcomes with Poisson or log-linear regression
- [Instructor] Standard models of regression are designed for continuous outcomes that measure on a scale of some kind, like an average. But when what you are trying to model is frequencies or how many cases fall into a category, you need to use a different model, and the most common approach goes by two different names. It's either called Poisson regression or log-linear models. They're the same thing. And I want to demonstrate this by first loading up R packages. And then I'm going to come down and use an example dataset that's in R called insect sprays. Let's get a little bit of information about that one. And what I like about this is that it's just oracle. It's from a paper published in 1942 about just about six different kinds of insect sprays and how many bugs they each killed, that's what's called count. Let's save that data, insect sprays, to DF for data frame, and let's take a quick look at the first few rows.…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Contents
-
-
-
-
-
-
Predicting outcomes with linear regression8m 49s
-
Predicting outcomes with lasso regression7m 48s
-
Predicting outcomes with quantile regression6m 27s
-
Predicting outcomes with logistic regression12m 49s
-
Predicting outcomes with Poisson or log-linear regression3m 43s
-
Assessing predictions with blocked-entry models10m 35s
-
-
-