Influence of butter and margarine prices on the demand for butter in Belgium
比较了三种家庭消费模型，发现Swamy随机系数回归模型优于固定系数模型，并指出黄油被视为奢侈品，但不同地区和社会阶层存在显著差异。
This paper discusses three types of models to describe household consumption behaviour using time series of cross section data. Swamy's Random Coefficient Regression Model is shown to be superior to a Fixed Coefficient Model if we are to obtain a macro relation explaining butter consumption. The Ordinary Least Squares Model is rejected on the basis of the data. The regression results give a strong indication that butter is considered a luxury good, although significant differences between regions and social classes exist.