This paper tests the predictive performance of machine learning methods in estimating the illiquidity of US corporate bonds. Machine learning techniques outperform the historical illiquidity-based approach, the most commonly applied benchmark in practice, from both a statistical and an economic perspective. Gradient-boosted regression trees perform particularly well. Historical illiquidity is the most important single predictor variable, but several fundamental and return- as well as risk-based covariates also possess predictive power. Capturing nonlinear effects and interactions among these predictors further enhances forecasting performance. For practitioners, the choice of the appropriate machine learning model depends on the specific application.