Econometric analysis of factors influencing the consumption of alcohol and cigarette

Regression models examines the relationship between the response variable and explanatory variables. For the use of Linear regression model which has a wide use, response variable needs to be continuous. In field of application this assumption is mostly not satisfied because of the categorical data usage. Thus, qualitative choice models are applied frequently in application. When the dependent variable is categorical, the ordinary least squares method can no longer produce the best linear unbiased estimator.In this study, throughout Turkey and 7 region, the factors effective on alcohol and cigarette expenditure are estimated and interpreted through ordered logit models. Household budget surveys data which has been carrying out by Turkish Statistical Institute was used at the research. According to the research results, throughout Turkey, monthly income, gender, age, educational level, marital status, business status, business standing, scope of business, dining out, playing game of chance, going coffee house, having private insurance and type of house factors effect alcohol expenditure; monthly income, gender, age, having health insurance, educational level, marital status, business standing, scope of business, dining out, playing game of chance, going coffee house, settlement type, occupation, property factors effect cigarette expenditure. Furthermore, aforementioned variables effects on alcohol and cigarette expenditure for 7 region are investigated.Keywords; 1) Alcohol Consumption, 2) Cigarette Consumption, 3) Ordered Logit Model, 4) Qualitative Choice Models