In Part I, the digital filter weak constraint four-dimensional variational (4D-Var) data assimilation is implemented with different digital filter weight settings through two typhoon cases. The results show that the influence of the digital filter weight on typhoon track and intensity prediction is large, and how to choose an appropriate digital filter weight is important. In this paper, seven different initial time typhoon cases in 2010 are selected to analyze the effects of different digital filter weight settings on typhoon track and intensity prediction. Based on the track and intensity prediction information of the seven cases, and the three-hour pressure tendency information for the whole domain and the area with 800 km radius from the typhoon center, a method of choosing digital filter weight is proposed for typhoon initialization with bogus data assimilation. Also, the ordinary digital filter weight setting is proposed, which is 1000 and obviously different from the operational and storm cases weight setting.