Sm. Nusser et al., A SEMIPARAMETRIC TRANSFORMATION APPROACH TO ESTIMATING USUAL DAILY INTAKE DISTRIBUTIONS, Journal of the American Statistical Association, 91(436), 1996, pp. 1440-1449
The distribution of usual intakes of dietary components is important t
o individuals formulating food policy and to persons designing nutriti
on education programs. The usual intake of a dietary component for a p
erson is the long-run average of daily intakes of that component for t
hat person. Because it is impossible to directly observe usual intake
for an individual, it is necessary to develop an estimator of the dist
ribution of usual intakes based on a sample of individuals with a smal
l number of daily observations on a subsample of the individuals. Dail
y intake data for individuals are nonnegative and often very skewed. A
lso, there is large day-to-day variation relative to the individual-to
-individual variation, and the within-individual variance is correlate
d with the individual means. We suggest a methodology for estimating u
sual intake distributions that allows for varying degrees of departure
from normality and recognizes the measurement error associated with o
ne-day dietary intakes. The estimation method contains four steps. Fir
st, the original data are standardized by adjusting for nuisance effec
ts, such as day-of-week and interview sequence. Second, the daily inta
ke data are transformed to normality using a combination of power and
grafted polynomial transformations. Third. using a normal components-o
f-variance model, the distribution of usual intakes is constructed for
the transformed data. Finally, a transformation of the normal usual i
ntake distribution to the original scale is defined. The approach is a
pplied to data from;he 1985 Continuing Survey of Food Intakes by Indiv
iduals and works well for a set of dietary components that are consume
d nearly daily and exhibit varying distributional shapes.