This study develops a model of auto insurance demand using logit models, based on risk aversion measure along with other individuals' characteristics such as age, sex, marital status, occupation, education, usage, engine size, number of accidents, driving experience and age of the car. Data for this study was collected during 2012, by random sampling methods and questionnaire. Using logistic regression model and With 497 completed questionnaires our results show that risk aversion has a significant and positive relationship with the auto insurance demand. Also marital status, governmental job, education and number of accidents variables have a significant and positive relationship with the auto insurance demand and sex and age of the car variables have a significant and negative relationship with the auto insurance demand.