Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
A recent demand system (AIDS) is extended to include family size explicitly and, then, estimated on Indian budget data. The estimation, using non-linear FIML, is performed, first, on pure time series and, then, on pooled cross section data. Within and across equation restrictions are imposed and tested. The principal results include:u(a)significant price and family size effects on the budget share of an item,(b)sensitivity of test results on homogeneity and symmetry to (i) rural/urban data, (ii) time series/cross section and (iii) inclusion/omission of family size,(c)considerable difference between pure time series and pooled cross section estimates, and(d)difference between the expenditure responses of the rural and urban consumer.