PARAMETER VARIATION AND NEW PRODUCT DIFFUSION

In this paper, an empirical investigation into parameter variation in diffusion models is conducted. Specifically, parameter estimates for t wo consumer durable products are obtained for time-invariant, flexible -form and stochastic-parameter specifications. Existing diffusion mode ls considered in the empirical analysis include the Bass (1969), Easin gwood, Mahajan and Muller (1983), Kamakura and Barasubramanian (1987) and Horsky (1990) diffusion models. In addition, a new model is develo ped that can be estimated with varying parameter structures, and which includes marketing-mix variables and replacement sales. In the empiri cal analysis, three estimation procedures are employed: non-linear lea st squares, a stationary stochastic procedure (Harvey and Phillips' 19 82 'Return to Normality' model using the Kalman filter), and a non-sta tionary stochastic specification (Cooley and Prescott, 1973, 1976). Th e results suggest that stochastic parameter specifications can be easi ly used to produce substantially better fits and that the improvement can be dramatic. Stochastic parameter specifications are especially us eful in the case of weak priors on the likely pattern of variation. Si nce some degree of parameter variation is often likely to exist, speci fying the exact form of the variation is important, albeit difficult. Stochastic parameter specifications can be very helpful in this regard . In addition, tracing the parameter path over time can assist in dete cting how current period parameter estimates deviate from the average over this life of the sample. (C) 1998 John Wiley & Sons, Ltd.