英文文献：Bayesian Estimation of The Impacts of Food Safety Information on Household Demand for Meat and Poultry-食品安全信息对家庭肉类和家禽需求影响的贝叶斯估计
英文文献作者：Taylor, Mykel R.,Phaneuf, Daniel J.
英文文献摘要：
Consumer reaction to changes in the amount of food safety information on beef, pork, and poultry available in the media is the focus of this study. Specifically, any differences in consumer reactions due to heterogeneous household characteristics are investigated. The data used in this study are monthly data from the Nielsen Homescan panel and cover the time period January 1998 to December 2005. These panel data contain information on household purchases of fresh meat and poultry as well as demographic characteristics of the participating households. The data used to describe food safety information were obtained from searches of newspapers using the Lexis-Nexis academic search engine. Consumer reactions are modeled in this study using a demand system that allows for both discrete and continuous choice situations. A seemingly unrelated regression (SUR) tobit model is estimated using a Gibbs sampler with data augmentation. A component error structure (random effects model) is incorporated into the SUR tobit model to account for unobserved heterogeneity of households making repeated purchases over time. Estimates of food safety elasticities calculated from the random effects SUR tobit model suggest that food safety information does not have a statistically or economically significant effect on household purchases of meat and poultry.

消费者对媒体上有关牛肉、猪肉和家禽的食品安全信息数量变化的反应是本研究的重点。具体地说，任何消费者反应的差异由于异质的家庭特征被调查。在这个研究中使用的数据是每月数据从尼尔森家庭面板和涵盖的时间周期1998年1月到2005年12月。这些面板数据包括家庭购买新鲜肉类和家禽的信息以及参与家庭的人口特征。用于描述食品安全信息的数据是从使用Lexis-Nexis学术搜索引擎的报纸搜索中获得的。消费者的反应在本研究中被建模使用一个需求系统，允许离散和连续的选择情况。利用吉布斯采样器进行数据扩充，估计了一个看似不相关的回归(SUR) tobit模型。一个组成误差结构(随机效应模型)被纳入SUR tobit模型，以解释在一段时间内进行重复购买的家庭未观察到的异质性。根据托比特随机效应模型计算出的食品安全弹性的估计表明，食品安全信息对家庭购买肉类和家禽没有统计上或经济上显著的影响。