Signal extraction and estimation of a trend: A Monte Carlo study

Several authors (King and Rebelo, 1993; Cogley and Nason, 1995) have questi oned the use of exponentially weighted moving average filters such as the H odrick-Prescott filter in decomposing a series into a trend and cycle, clai ming that they lead to the observation of spurious or induced cycles and to misinterpretation of stylized facts. However, little has been done to prop ose different methods of estimation or other ways of defining trend extract ion. This paper has two main contributions. First, we suggest that the deco mposition between the trend and cycle has not been done in an appropriate w ay. Second, we argue for a general to specific approach based on a more gen eral filter, the stochastic trend model, that allows us to estimate all the parameters of the model rather than fixing them arbitrarily, as is done wi th mainly of the commonly used filters. We illustrate the properties of the proposed technique relative to the conventional ones by employing a Monte Carlo study. Copyright (C) 1999 John Wiley & Sons, Ltd.