Assumptions of the probit model
1. The relationship between the dependent variable and the independent variables is linear in the probit model.
2. The error terms in the probit model are normally distributed.
3. The independent variables are not highly correlated with each other (i.e., there is no multicollinearity).
4. The error terms are homoscedastic (have constant variance).
5. The observations are independent of each other.
6. The dependent variable is binary or dichotomous in nature (i.e., it can only take on two possible values).
7. The probit function is used to model the probability of the dependent variable taking on a particular value.