At the present research stage, quantitative investment is mainly aimed at investment products with small fluctuations, while the benefits in markets with large fluctuations are not significant. In view of this, this paper takes gold and bitcoin as examples to establish a quantitative investment selection model based on multiple indicators. The model first uses XGBoost to predict the price of the next day. At the same time, Prediction Based Mean‐Variable Portfolio Optimization model (PBMVPO) is established to calculate the position ratio between bitcoin and gold, and reasonably complete the risk balance between high‐risk products and low‐risk products. Then the transaction strategy is generated according to the evaluation results. Finally, in order to evaluate the effectiveness of the model, the model is used to back test the data from 2013 to 2016. In addition, the robustness analysis of the model is carried out to verify the robustness of the model.