INDIVIDUAL DEMANDS FROM HOUSEHOLD AGGREGATES - TIME AND AGE VARIATIONIN THE COMPOSITION OF DIET

This paper uses non-parametric regression with smoothness induced by i mposition of a roughness penalty to estimate time-dependent relationsh ips between age and intakes of energy, total fat and saturated fatty a cids employing data on household food acquisitions from a continuous s urvey covering the period 1975-94. A discrete analogue of the bivariat e thin plate spline smoother is used to smooth the estimates across ag e and time. With age and calendar time-specific estimated intakes in h and it is possible to track changes with age within cohorts. Of partic ular interest in this application are estimates of the change through time and across age in measures of dietary composition which figure in many governments' dietary targets such as the proportion of energy de rived from fat. The method used here can be applied in any situation i n which household aggregate data are available on consumption of priva te goods and there is a need to model the demands of household members or to control for household composition, opening the way to the const ruction of individual pseudo-panels from household aggregates. (C) 199 8 John Wiley & Sons, Ltd.