The results of water budget closure-based evaluation using budget closure residuals showed different results with the statistical methods. The budget closure residual showed that GLDAS and FLDAS had the best accuracy of water budget closure, followed by GPM-IMERG, TRMM-3B43 and PERSIANN-CDR. This fully illustrates the importance of evaluating HDPs from these two perspectives. Because these two kinds of methods guarantee data accuracy of HDPs from different dimensions, i.e. the data error and budget closure error. Both dimensions are important for hydrological analysis. However, water budget closure-based methods showed some advantages compared with statistical methods. For example, the error of water budget-based method has temporal variation distribution because they are calculated based on water budget equation at each time scale such as a month or a year. This determined that the results of this kind of methods can used to analysis the monthly, seasonal and yearly changes of water budget errors. More importantly, this kind of methods can calculate errors of HDPs in ungauged regions and can effectively avoid false evaluation results caused by information overlap. This kind of methods greatly compensates for the lack of gauge-based evaluation methods.