摘要：A Cox nonnested test with parametric bootstrap is developed to select between thelinearized version of the First Difference Almost Ideal Demand System(FDAIDS) and the Rotterdam model. The Cox test with parametric bootstrap isexpected to be more powerful than the various orthodox tests used in pastresearch. The new approach is then used for U. S. meat demand (beef, pork, andchicken) and compared to results obtained with an orthodox test. The orthodoxtest gives inconsistent results depending on the inclusion or exclusion of fish andthe time period covered. In contrast, under the same varied conditions, the Coxtest with parametric bootstrap consistently indicates that the Rotterdam model ispreferred to the FDAIDS.
