Exposure to fine particulate matter (PM_2.5) remains a worldwide public health issue. However, epidemiological studies on the chronic health impacts of PM_2.5 in the developing countries are hindered by the lack of monitoring data. Despite the recent development of using satellite remote sensing to predict ground-level PM_2.5 concentrations in China, methods for generating reliable historical PM_2.5 exposure, especially prior to the construction of PM_2.5 monitoring network in 2013, are still very rare. In this study, a high-performance machine-learning model was developed directly at monthly level to estimate PM_2.5 levels in North China Plain. We developed a random forest model using the latest Multi-angle implementation of atmospheric correction (MAIAC) aerosol optical depth (AOD), meteorological parameters, land cover and ground PM_2.5 measurements from 2013 to 2015. A multiple imputation method was applied to fill the missing values of AOD. We used 10-fold cross-validation (CV) to evaluate model performance and a separate time period, January 2016 to December 2016, was used to validate our model's capability of predicting historical PM_2.5 concentrations. The overall model CV R² and relative prediction error (RPE) were 0.88 and 18.7%, respectively. Validation results beyond the modeling period (2013–2015) shown that this model can accurately predict historical PM_2.5 concentrations at the monthly (R² = 0.74, RPE = 27.6%), seasonal (R² = 0.78, RPE = 21.2%) and annual (R² = 0.76, RPE = 16.9%) level. The annual mean predicted PM_2.5 concentrations from 2013 to 2016 in our study domain was 67.7 μg/m3 and Southern Hebei, Western Shandong and Northern Henan were the most polluted areas. Using this computationally efficient, monthly and high-resolution model, we can provide reliable historical PM_2.5 concentrations for epidemiological studies on PM_2.5 health effects in China.