An Alternative Approach to the American Demand for Money
用Hendry（1979）的方法重新分析1973年后美国货币需求的结构性断裂，检验是否存在真正的需求函数变化，并解释为何其他研究者得出存在断裂的结论。
WITH A DECADE OF EVIDENCE BEHIND IT, the profession seems to have agreed on the fact that a structural break shifted the American demand for shortly after 1973. It is difficult to pinpoint the source of the trouble precisely, let alone to explain the event with an econometric model. In their survey, Judd and Scadding (1982) tentatively concluded that financial innovation is the source of the problem. While this conclusion has not been universally accepted, the instability of the demand function is largely taken as a given stylized fact. The primary goal of this article is to provide an alternative solution to the case of the missing money using the methods of Hendry (1979). It should be stressed at the outset that, as this paper deals with a problem in the applied literature, it is unabashedly empirical in nature; the paper attempts to explain an empirical difficulty without presenting a new theoretical model of demand. Rather, in the context of the standard transactions demand theory for narrow money, a number of questions are analyzed with an atypical econometric model. Was there actually a shift, or a series of shifts in the demand for money? If not, why have other investigators concluded that there was in fact a shift? Most importantly, what function seems to be the most reasonable approximation to the data generation process during the period in question?