Converging conflicting forecasting parameters in forecasting durable new product sales

Forecasting sales for a durable product prior to the product's introduction is a necessary but difficult task. Most new product forecasting models require the estimation of parameters for use in the forecasting model. For a specific parameter the forecaster is often faced with estimated parameter values that are not in agreement. Dealing with this problem is the subject of this paper. Specifically, a methodology of dealing with conflicting parameter values is applied to a diffusion forecasting model. Diffusion models are widely used to forecast new product sales for durable products.