Beverage consumption among US children and adolescents: full-information and quasi maximum-likelihood estimation of a censored system

Milk, soft drink and juice consumption is investigated for children and adolescents in the USA. Full-information maximum likelihood estimator (FIML) and 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 by soft drinks as a child or adolescent grows older. Income, TV watching, gender, race, and other demographic variables also play significant roles in determining beverage consumption.