Calculation of decision making probability using probit and logit models

The aim of this article is presentation of logit and probit models and their wide application in many different science. Logit and probit regression are used for analyzing the relationship between one or more independent variables with categorical dependent variable. There are a lot of advantages of logit (probit) models over linear multiple regression. These methods imply that the dependent variable is actually the result of a transformation of an underlying variable, which is not restricted in range. For example, the probit model assumes that the actual underlying depedent variable is measured in terms of values for normal curve; if one transforms those values for probabilities then the predictions for the dependent variable will always fall between 0 ond 1. Thus, we are actually predicting probabilities from the independent variables The probit model was used to calculate the probability of admissions in Rzeszow Uniwersity, speciality Handel i spoldzielczosc.