Probit Regression Calculator

Probit regression, also called the probit model, is used to for the modelling of dichotomous or binary outcome variables (0 or 1). The inverse standard normal distribution of the probability is modeled as a linear combination of the predictors by transforming sigmoid-response curves to a straight line that can be analyzed by non-linear regression with techniques like least squares or maximum likelihood estimator. Probit analysis is primarily used in the understanding of dose-response relationships but has applications in other fields as well. The statistical analysis generates estimations for beta coefficients, p values, standard errors, log likelihood, residual deviance, null deviance, and AIC, which can be used in making predictions about dichotomic responses.