The presentation of a theoretically sound and consistent methodology that could help policymakers to evaluate the welfare effects of alternative price policies was the focus of this Dissertation work. The proposed methodology had three basic stages: (1) classification of households in income groups; (2) estimation of demand systems for each of the newly formed income groups; and (3) measurement of welfare changes by means of estimating compensating variation measures from the underlying cost functions;The classification of households in income groups was based on an indepth understanding of the behavior of households as it relates to the acquisition of goods. Technically speaking, the methodology to classify households was based on an analysis of homoskedasticity of variances of residuals from regressions of Engel relationships. Four income groups were established. The Almost Ideal Demand System (AIDS) was used in this study to estimate the structural parameters of the demand equations for each income group. Household characteristics were incorporated by the demographic translating technique. The proportion of observations reporting zero expenditures for some food groups conditioned the methodology for the estimation of the AIDS models. Endogenous switching regressions techniques were used to get unbiased and consistent AIDS demand parameter estimates for the low income group. Standard seemingly unrelated equation techniques were used to estimate the AIDS demand parameters for the medium-low, medium-high and high income groups;The final stage of the proposed methodology implied the estimation of compensating variation measures. The following main policy implications were drawn from a simulation analysis of single and multiple price changes: the most appropriate policy was a direct transfer scheme to low income households only; reductions of direct and indirect price subsidies for meats and dairy products affected minimally low income households; increases in price of fish affected mainly low income households.