Choosing between the AIDS and Rotterdam models: A meat demand analysis case study

Due to the inability of economic theory to choose ex ante between the Almost Ideal Demand System (AIDS) and the Rotterdam model, a non-nested test was used. The results of the non-nested test points to the Linearized-AIDS model applied to 31 years of meat consumption data in South Africa. When comparing the estimated demand relations of the two models, the LA/AIDS model also proved to be better fit for South African meat demand.