So-called reanalysis data are useful for climatic data analysis. However, some reanalysis data, such as APHRODITE (Asia Precipitation Highly Resolved Observational Data Integration Towards Evaluation), have biases in their mean and variance, especially for extreme rainfall. Undeniably, it needs further analysis for improvement. This study applies a multivariate bias- correction method to Korea, southern China, Philippines, and Thailand to realize this improvement. Our focus in this study is the annual maximum daily precipitation (AMP1). After correcting APHRODITE data over grids, we construct a time series of the AMP1 observations for some locations with no observational records by applying the so-called back- interpolation technique. This technique is also used to estimate missing values in the AMP1 series. This method is illustrated in some cities in North Korea and the Philippines.