Early warning of bond default risk is to predict the future bond default status based on the enterprise’s financial factors, nonfinancial factors, and external macro factors. For different combinations of variables, the effect of default prediction is different; there must be an optimal combination of indicators, which can minimize the error of default prediction. For different thresholds, the effect of default prediction is different, and there is bound to be an optimal threshold of default judgment, which can distinguish between the bonds that default from those that do not to the greatest extent. This paper uses the random forest model to select the feature combination, and studies the default risk of bonds based on Logit model. The first contribution of this paper is in the optimal feature selection. Under the premise of the minimum 〖WTBX〗TypeII Error〖WTBZ〗, an optimal random forest is obtained by maximizing the 〖WTBX〗AUC〖WTBZ〗 for different numbers of decision trees. According to the ranking of the importance of indicators, the optimal feature combination can be obtained by forward selection to maximize 〖WTBX〗AUC〖WTBZ〗. The second contribution is in the determination of the optimal default judgment threshold. Taking the minimum weighted sum of the 〖WTBX〗TypeI Error〖WTBZ〗 and 〖WTBX〗TypeII Error〖WTBZ〗 as the objective function, the optimal threshold of logical regression is deduced. The third is the prediction accuracy of this model is higher than popular big data prediction models. Based on data from bonds issued by Chinese bonds listed companies from 2014 to 2018, the empirical research shows that the key indicators affecting China’s medium and shortterm default prediction are： Monetary capital / shortterm debt, net profit, the number of bonds issued by issuers, industry prosperity index, and industry entrepreneur confidence index. The key indicators affecting shortterm default prediction are： Monetary assets, quick ratio, fixed asset investment price index, and money supply 〖WTBX〗M〖WTBZ〗0. The key indicators that have an impact on the mediumterm default forecast are registered capital of the issuer, repayment amount of bonds at maturity, and bond maturity index.