Qualitative choice models, multinomial logit, probit models and an application

Dummy variables that take two or more values may be dependent variable in the regressionmodels. Qualitative choice models, where the dependent variable is qualitative, are separatedtwo groups as binary and multinomial. The basis aim in two groups of models is to determinethe choice probabilities.Binary choice models assume that decision makers are faced with a choice between twoalternatives and that their choice depends on their characteristics. The difference betweenprobit and logit models is that while probit is related to the normal distribution, logit is relatedto the logistic distribution.Decision makers are faced with more than two alternatives in multinomial choice models.Categorical variables are classified into three groups as ordered, unordered and sequential.The various multinomial choice models are formed by using these categorical variables. Therandom utility model is utilized to explain the choices of the decision makers. The randomutility of an alternative is expressed as a sum of observable or systematic and unobservablecomponents. The choice probability of alternative i is equal to the probability that the utilityof alternative i , U in , is greater than or equal to the utilities of all other alternatives in thechoice set. The error terms are assumed to be distributed multivariate-normally in themultinomial probit model while the error terms are assumed to be independently andidentically distributed as a Weibull distribution in the multinomial logit model.In this application, individuals were separated into five groups according to the working statusand working status were used as dependent variables. The ordered probit and ordered logitmodels were applied and for each group, the probabilities that were associated with the groupswere calculated.Keywords: Dummy variable, qualitative choice models, ordered probit, ordered logit,random utilityJURY:1. Ass. Prof. Dr. Ali Hakan BÜYÜKLÜ Date: 15.03.20062. Ass. Prof. Dr. Doğan YILDIZ Page: 1543. Ass. Prof. Dr. Osman Zeki ÇET NKAYA