Abstract
Domestic tire replacement demand is explained by a stochastic model using 3 distributions : car replacement, tire replacement and annual running distance distributions. At the first development, though the value of the 3 distributions are fixed throughout the range (fixed factor model), the results meet the real data very well. But, as the model treats a long term in which rapid social changes are involved, the difference between the model and the real data became significant, and adequate model revising in requested. The model is expanded so that the 3 distributions may vary reflecting the social changes (variable factor model), and this results in a reasonable forecasting of the model. This paper treats the stochastic model construction, estimation of the parameters, comparison of the simulation results with the real data and the revision process of the model.
