Beverage consumption among US children and adolescents: full-information and quasi maximum-likelihood estimation of a censored system
研究了美国儿童和青少年对牛奶、软饮料和果汁的消费，使用完全信息最大似然和准最大似然估计方法，发现随着年龄增长，软饮料替代了牛奶，收入、看电视、性别、种族等人口变量也显著影响饮料消费。
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. Copyright 2002, Oxford University Press.