Aim: The main objective of this study is to predict the Sales of E-commerce using Extreme gradient boosting which is a base model and compared with Arima model algorithm. Materials and Methods: Extreme gradient boosting and Arima model algorithms are used to predict the Sales of E-commerce. Sample size is calculated using G Power calculator and found to be 25 per group has been taken and a total of 50 samples are used. Where Pretest power is 80% and CI of 95%. Results: Based on the analysis Extreme gradient boosting has significantly more accuracy (83.50) compared to random arima model algorithm (79.50). There is a Statistically Significant difference between the two groups with p=0.02 (p<0.05). Conclusion: According to this study Extreme gradient boosting has better accuracy than the Linear regression algorithm to predict the sales prices in E-commerce .