With the development of the used car market, the demand for a more accurate and scientific price prediction model of used cars becomes urgent. This paper uses multiple linear regression,decision tree and random forest to build up the automobile price forecasting model. We use means to cluster cars and find out that some factors like power, kilometers, gearbox have an influence on the price. According to the analysis, we find out that random forest has the best prediction performance,make sure R2 reaches 0.92 will be enough.