Empirical Measurement of an inflation Index: A Multiple-Indicators Distributed-Lag Approach
扩展了多原因模型，引入几何分布滞后处理不可观测变量，用最大似然估计法测量批发层面的通胀指数，并证明增加指标数量可提高估计效率。
Zellner (1970) considered an interesting model containing an unobservable variable that was treated as a linear combination of many observable variables. Goldberger (1972a,b, 1974) extended Zellner's model to the case of several dependent variables in the multiple causes model (MCM) and the multiple indicator multiple cause model. This article presents a dynamic extension of the MCM that allows for a geometrically distributed lag in the unobservable variable. The maximum likelihood estimation technique for the parameters of the model is derived. The model is then used to obtain the empirical measurement of the inflation index at the wholesale level. Furthermore, it is shown that under certain conditions an increased number of indicators leads to a more efficient estimator.