Robertson, J. C.2016-08-282016-08-281987https://hdl.handle.net/10182/7281Several theoretically derived competing explanations of consumer demand behaviour in New Zealand are considered as an illustration of the use of econometric model-encompassing techniques in the theory-discrimination process. The research strategy used incorporates both standard diagnostic checking procedures and model specification tests as well as criteria for testing nested and non-nested models. The principle result is that while the Almost Ideal Demand System can withstand both evaluation in terms of its own performance as well as its ability to predict the performance of the alternative models, its own performance can not be similarly predicted by the others.enhttps://researcharchive.lincoln.ac.nz/pages/rightsconsumer demanddata-generating processdemand analysisdiscriminant analysiseconometricsencompassing principlehypotheses testingmodelsnested specification testsnon-nested specification testsnon-linear modellingThesisDigital thesis can be viewed by current staff and students of Lincoln University only. If you are the author of this item, please contact us if you wish to discuss making the full text publicly available.ANZSRC::140303 Economic Models and ForecastingANZSRC::140305 Time-Series AnalysisQ112848305