Answers are grounded in this paper and cite their sources.
A Rotterdam import allocation model is used to fit import data for fresh apples in four importing markets important to U.S. apple exporters. Nested tests rejected homotheticity but could not reject homogeneity, symmetry, or separability among import suppliers. A Monte Carlo test rejected first-order autocorrelation in each market. Expenditure and price elasticities are calculated and reported.
No takes yet. Share an insight, caveat, or question.
Seale et al. (1992) studied this question.