Diffusion of Technical Change and the Decomposition of Output into Trend and Cycle
认为将实际GNP趋势建模为随机游走与生产率增长的解释不一致，提出用S形脉冲响应的ARIMA模型替代，并基于美国战后季度数据估计，发现模型更简洁且拟合效果相当，周期/趋势方差比很低。
In this paper, the authors argue that modeling the trend component in real GNP as a random walk is inconsistent with its interpretation as productivity growth. As an alternative, they specify the trend as an ARIMA whose impulse response function follows an S-shaped pattern reflecting the process of diffusion of technical change. Such an ARIMA is employed to build and estimate an UCARIMA using U.S. postwar quarterly data. The authors find that their model, although more parsimonious, fits the data equally as well as the standard random walk plus AR(2) cycle. Moreover, their model has a very low cycle/trend variance ratio. Copyright 1994 by The Review of Economic Studies Limited.