Multi-source weighted-ensemble precipitation (MSWEP) is a high-resolution gridded precipitation product, in this study its simulation accuracy for the precipitation in the coastal area of China’s mainland is evaluated in detail. This evaluation is based on the precipitation data collected from 96 meteorological stations located in the coastal provinces from 1979 to 2020. In this study, we adopted Cressman algorithm to correct the MSWEP, analysed the sensitivity of the search radius (R) to the correction results, and evaluated the effectiveness of the corrected MSWEP for precipitation simulation. The results show that MSWEP tends to underestimate the daily precipitation across the entire coastal region, with a slightly better performance for the northern provinces than the southern provinces and significantly underestimated results in extreme precipitation period. During the typhoon rainstorm period, the overall data error is unacceptably large and the underestimation becomes more obvious with increasing precipitation. According to the sensitivity analysis of R, the optimal value of R is 2.0°and applying correction three times yields the best results. The overall performance of the corrected MSWEP for precipitation simulation is enhanced, and its accuracy for extreme rainstorm simulation is significantly improved. The research outcomes can provide a data support for precipitation related research, as well as disaster prevention and mitigation in the coastal area of China.