1

I have run probit and logit regression analysis for a model, where the dependent variable is 'Purchase' which takes the value of 1 for purchase, else zero. The independent variable is the log of distance from the highway. The unit of distance from the highway is in miles. I have calculated the marginal effects: enter image description here

I have interpreted the effect as:

A one-mile increase in travel distance from the highway increases the chance of purchase by 0.0099Xln(1.01) = 0.000098 percent (probit regression).

Does my interpretation look correct? If not, how I should explain?

Jui Sen
  • 111
  • 1
    Logit case is covered here and probit case is here. – dimitriy Dec 21 '22 at 21:18
  • 1
    You can interpret that as saying that a 1% increase in distance is associated with a 0.0099% increase in the probability of purchase (on a [0,100] scale). That's 1/100th of a percent. You can also do it in terms of a 10% increase in distance, which gets you 1/10 of a percent. – dimitriy Dec 21 '22 at 21:25

0 Answers0