A probit model for beer demand
You heard other data scientists prefer using the probit response function for modeling purchase decisions. The probit treats purchases decisions as latent propensities. That sounds fancy and makes you nervous, so you try out the probit, too.
You can again use the function glm() to describe the relation HOPPINESS ~ price.ratio. You only need to augment the family argument by binomial(link = probit). As usual, the estimated coefficients are obtained by using the function coef().
Cet exercice fait partie du cours
Building Response Models in R
Instructions
- Explain
HOPPINESSbyprice.ratiousing the functionglm()and the argumentfamily = binomial(link = probit). Assign the result to an object namedprobit.model. - Obtain the coefficients of the
probit.modelby using the functioncoef().
Exercice interactif pratique
Essayez cet exercice en complétant cet exemple de code.
# Explain HOPPINESS by price.ratio
probit.model <- ___(___, family = ___, data = choice.data)
# Obtain the coefficients