Indicators for Dating Business Cycles: Cross-History Selection and Comparisons
比较了两种确定商业周期转折点的方法：先汇总再定时与先定时再汇总，利用270个月度经济指标提供初步证据，对经济周期研究者有参考价值。
(CEPR) Business Cycle Dating Committee date business cycle turning points using a small number of aggregate measures of real economic activity. For example, in its memorandum explaining the December 2007 peak (NBER Business Cycle Dating Committee 2008), the NBER committee mentioned that it considers five series, quarterly real GDP and the “big four ” monthly series, real personal income less transfers, real manufacturing and wholesaleretail trade sales, industrial production, and nonfarm employment. (These series do not in general receive equal weight.) In contrast, when the NBER research program on dating business cycles commenced, researchers examined turning points in hundreds of series and dated business cycles by detecting clusters of specific-cycle turning points; see Arthur Burns and Wesley Mitchell (1946, 13 and 77–80). The dating of turning points evidently has shifted from aggregating the turning points of many disaggregated series to using the turning points of a few highly aggregated series. This shift raises a methodological question: should reference cycle turning points be determined by aggregating then dating, or by dating then aggregating? This paper provides some preliminary evidence on the question of whether it is better to date then aggregate or aggregate then date using 270 monthly disaggregated real economic indicators. The questions considered in this paper parallel those in the large literature on forecasting