本研究將RFM模式加以擴充，加入了二個變數： T (Time)表示顧客距第一次購 買的時間、C (Chum) 表示顧客每歷經一次行銷後流失的機率，再利用機率理論中的 Bernoulli序列，導出可以預測顧客購買次數的期望值、顧客終生回應次數期望值、顧客行銷價值、顧客終身價值的模型。本研究也提出估計模型參數的方法，並以一個實際個案進行實證研究。研究結果顯示，本研究所導出的理論模型可以準確預測顧客未來購買次數的期望值。

This study expanded the RFM model by including two parameters, time since first purchase (T) and chum probability (C). Using Bernoulli sequence in probability theory, this study derived out the model that can predict the expected value of the number of times that the customer will buy in the future, customer's marketing value at the next campaign, and customer lifetime value. This study also proposed the methodology to estimate the unknown parameters, such as chum probability, in the formulas, and examined a real case to verify the model. The findings showed that the model can accurately predict the expected value of the number of times that the customer will buy in the future.