We perform a validity test of a long-term replenishment contract model proposed by previous research. The model is unique in that demand is assumed to be generated from an ARIMA process. For practical implementation, it approximates shortage cost to build expected cost of the contract policy. Therefore, it is worthwhile to test the validity of the model. We check the difference of the values obtained from the model with theoretical ones. The result shows that the model has a satisfactory degree of accuracy.