Türkiye Hanehalkı Eğlence Kültür Harcamalarında Tobit Modelin Kullanımı

Social scientists often meet ccnsorcd, truneatcd or limited dcpcntcnt variables. The limited structure of these variables is observed in the wages' being absolutely not negative for instance, or the example process obtained with the truncation of wage distribution of those working for minimum wage in the population. Limited dependent variables arc not symmetrical but skewed and this skewness may be the unsymmctrical result of ccnsorcd or truncated variable. Besides, the present skewness in the variables can also be rcflactcd. However, the skewness conccmcd in the variables is ncglcctcd in many studies, Censored or truncated regression models arc known as Tobit model. Bccausc the incomc of the family members is low for instance, there may be no expenses for entertainment or culture until a certain level is reached, or there may be no labour supply until the wage rate reaches a certain limit. Tobit models refer to regression model with dependent variables which have lower and upper limits. In a standart linear regression model, even when the residuals' normality assumption is violated, the OLS estimators arc unbiased and consistent. Tobit model estimates arc not resistant to such violations.The aim of this study is to attract the attention to the OLS use problems after continues or ordinal dependent variables, using the entertainment and culture expenses of the households (12640 data) the incomc, sex, age, éducation level, marital status and information about the jobs of household heads benefiting from the consumption data of the Turkish households in 1994, and to try to explain the connection between Tobit and Probit model by comparing it with the traditional OLS estimator.