Abstract: Based on Baidu index and taking Yu’E Bao asset volume as an example, market demand prediction models containing Internet search volume and excluding Internet search volume are constructed and compared by using ARIMA model Comparative study shows that prediction model containing search volume has high precision, which means the data contains some factors not considered in traditional model The introduction of this variable is beneficial to improving the prediction precision of the model, providing more accurate results for the product or service market decision-making

Key words: Internet search volume, Baidu index, asset size of Yu’e Bao, prediction model