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.
Deze oefening maakt deel uit van de cursus
Building Response Models in R
Oefeninstructies
- Explain
HOPPINESSbyprice.ratiousing the functionglm()and the argumentfamily = binomial. Assign the result to an object namedlogistic.model. - Obtain the coefficients of the
logistic.modelby using the functioncoef().
Praktische interactieve oefening
Probeer deze oefening eens door deze voorbeeldcode in te vullen.
# Explain HOPPINESS by price.ratio
___ <- ___(___, family = ___, data = choice.data)
# Obtain the coefficients