Goodness-of-f it testing in growth curve models: A general approach based on finite differences

摘要

The method is general in that it is not limited to specific parametric forms underlying the hypothesized model so long as an appropriate finite difference of some function of size variables can be made to vanish. Also it allows testing to de done under a set up which manages to relax most of the assumptions made in Ref. 1, which are generally reasonable but not guaranteed to hold universally. The performance of the theory developed is illustrated using simulated and real life data. (16 refs.)