The Effects of Segmenting Quarterly Sales and Margins on Extrapolative Forecasts of Conglomerate Earnings: Extension and Replication
复制并扩展了Kinney和Collins的研究，利用季度数据检验分割利润信息是否提升收益预测准确性，发现分割销售数据已足够，利润信息增益有限。
Several years ago, Kinney [1971] and Collins [1976] compared income forecasts based on consolidated (CN) data to income forecasts based on segmented (SG) data. Somewhat paradoxically, their results imply that SG profit margins may not add very much to the accuracy of income forecasts, once SG sales are considered. Indeed, only nominal improvements were observed when using SG instead of CN margins. Their findings, however, were limited to some extent by data constraints. Segment reporting was new at the time and data sufficient for time-series analysis were not available. This prevented them from using the same forecasting techniques on both the CN and the SG data sets, a limitation recognized by Collins [1976, p. 167]. Moreover, the data were subject to problems of segment definition, transfer pricing, common cost allocations, and even earnings definition, thereby leading to a potential lack of comparability across firms (Collins [1976, p. 164]).1 My purpose here is to extend to quarterly reporting the studies of Kinney and Collins with regard to the predictive benefits of SG margins.