Why you should never use the Hodrick-Prescott Filter: Comment

Abstract. Hamilton (2018) argues that one should never use the Hodrick-Prescott (HP) filter, given its drawbacks and the existence of a better alternative. This comment shows that the main drawback Hamilton finds in the HP filter, the presence of filter -induced dynamics in the estimate of the cyclical component, is also a key feature of the alternative filter proposed by Hamilton. As with the HP filter, the Hamilton filter applied to a random walk extracts a cyclical component that is highly predictable, that can predict other variables, and whose properties reflect as much the filter as the underlying data-generating process. In addition, the Hamilton trend lags the data by construction and there is some arbitrariness in the choice of a key parameter defining the filter. Therefore, a more balanced assessment is that the HP and Hamilton filters provide different ways to look at the data, with neither being clearly superior from a practical perspective.

JEL Codes: B41, C22, E32.

Keywords: HP filter; Hamilton filter; business cycles; detrending; filtering.