The Basics
- Simple definition: A statistical model for binary outcomes (yes/no) using the normal distribution.
- Core idea: Estimates the probability of an event using a normal cumulative distribution function.
- Think of it as: Logit’s close cousin – same purpose, slightly different mathematical assumption.
What It Actually Means
Like logit, probit models handle situations where the dependent variable is binary. The difference is mathematical: probit uses the cumulative normal distribution while logit uses the logistic distribution. In practice, they usually give similar results. The choice often depends on discipline preference or convenience. Coefficients are interpreted through marginal effects – how much a one-unit change in a variable changes the probability of the outcome.
Example
Studying loan default in Pakistan’s banking sector, a researcher might use probit to estimate how factors like income, loan size, and business type affect default probability. The model predicts each borrower’s default risk.
Why It Matters
Probit models are standard in economics, finance, and public health for analyzing binary choices – from market entry decisions to vaccination take-up.
Don’t Confuse With
Logit Model – both do similar things; probit assumes normal distribution, logit assumes logistic. Results rarely differ meaningfully.
See also
Logit Model • Binary Choice • Econometrics • Marginal Effects • Maximum Likelihood Estimation
Read more about this with MASEconomics:
Logit and Probit Models: Understanding Binary Choice in Econometrics