Abstract
In long range trend projection, one of the most frequently used models is the “S”-shaped growth curve originally suggested by Pearl and Reed. Because of the nonlinear form of this model, problems exist in the statistical evaluation of model fit and in estimating the precision of forecasts developed from the model. Based on an approach first suggested by Yule, a procedure for fitting and evaluating the Pearl–Reed curve is developed. Confidence intervals for parameters and forecast prediction intervals are presented.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have