Key points are not available for this paper at this time.
ABSTRACT There has been much recent controversy concerning the interpretation of distance‐decay parameters estimated from gravity models. A relationship between spatial structure and estimated distance‐decay parameters has been posited. Empirical evidence is given here for such a relationship and theories attempting to explain the dependence of parameter estimates upon spatial structure are examined. These theories are shown to be weak and directions for future research are given.
A. Stewart Fotheringham (Tue,) studied this question.