Exercise

# A logistic model for beer demand

The linear model does not fit the data when it comes to predicting purchase probabilities. You need a response function that bounds the model predictions between zero and one.

The logistic response function can do this job for you. Therefore, you will need the function `glm()`

. The function `glm()`

works very similar to the `lm()`

function. The main difference is the additional `family`

argument. As `HOPPINESS`

is a binary variable, you have to specify the `family`

argument as `binomial`

.

Instructions

**100 XP**

- Explain
`HOPPINESS`

by`price.ratio`

using the function`glm()`

and the argument`family = binomial`

. Assign the result to an object named`logistic.model`

. - Obtain the coefficients of the
`logistic.model`

by using the function`coef()`

.