Beverage Consumption Among U.S. Children and Adolescents: Full-Information and Quasi Maximum-Likelihood Estimation of a Censored System

Steven T. Yen and Biing-Hwan Lin

Elsewhere No. (0221), August 2002

Juice, soft drink, and milk consumption are investigated using a sample of children and adolescents from the Continuing Survey of Food Intakes by Individuals. Full-information maximum likelihood (FIML) estimator and a parsimonious quasi maximum likelihood (QML) alternative are used to estimate a censored system of beverage equations. The QML estimator is found to be an acceptable alternative to the FIML estimator for the empirical application considered. We find displacement of milk and juice by soft drinks as a child or adolescent grows older. Income, TV watching, gender, race, and other demographic variables are also found to play significant roles in determining beverage consumption.

Keywords: Censoring, microdata, milk, soft drinks

Get this Document in Adobe Acrobat PDF format.

Updated date: August 2002