Nº 45 (January, 2007). Georgina Pizzolitto. 

This paper examines food consumption in Argentinean households through the estimation of food Engel curves. It also considers households demographic characteristics and heterogeneity in consumption preferences. The estimation of different functional forms for the Engel curves, using parametric and semiparametric techniques, suggests that both, non linear and Fourier flexible functional form are the best approximations to work with and adequate represent the data. Quantile regression confirms that the relation between the share of budget spend on food and the logarithm of household expenditure per head differs at different points in the conditional distribution. Household demographic characteristics are also important in determining the share of the household budget devoted to food and the consumption patterns.

JEL Code: D12, C14