Based on the sample data of agricultural soil profile, the locally-developed pedo-transfer function (PTF) of agricultural soil saturated hydraulic conductivity (Ks) in Daxing district was established using multiple regression with the independent variables, including constant and the monomials with variables of Si content, Cl content, and organic matter content. The PTF was validated simultaneously using multi-fold cross-validation method. The performance of Daxing District PTF built in this paper was compared with other three existing PTFs in predicting Ks. The PTF from this study was used in regional prediction of Ks in depth of 0~80 cm. The results showed that the root-mean-square error, average absolute relative error, and coefficient of determination of multiple regression function for predicting Ks in depth of 0~100 cm were 40.525 cm/d, 204.738%, 0.544, respectively, which were much lower than that of other three PTFs. The Ks was of great spatial variability in the top layer soil of 0~20 cm, and was relative small along the Yongding River in the north and southwest of Daxing district. The PTF from this study also performed well in the characterization of layered Cl loam in depth of 60~80 cm in Caiyu. It can be used in regional evaluation of Ks in Daxing district.