The accuracy of the stock prediction model is notaccurateenough

for there are problems such as overfitting and underfitting. Based onthe analysis of the existing stock prediction methods

this paper proposes astock prediction method based on Cart decision tree and boosting. Currentforecasting models have less consideration in data verticality

so we adds theaverage price in the past ten days and turnover rate to improve the accuracy ofthe stock prediction.The method is based on the Cart decision tree method. We use boosting toconcatenation multiple decision tree to build a stock model to solve the overfittingproblem. Theprediction experiment results show that this stock prediction has a good effectin the field of instruments and meters

and the meansquare error has beendecreased.