Why You Should Never Use the Hodrick-Prescott Filter

By James D. Hamilton. Review of Economics and Statistics, 2017

1,324 citations recorded by the source corpus.

Abstract Here’s why. (a) The Hodrick-Prescott (HP) filter introduces spurious dynamic relations that have no basis in the underlying data-generating process. (b) Filtered values at the end of the sample are very different from those in the middle and are also characterized by spurious dynamics. (c) A statistical formalization of the problem typically produces values for the smoothing parameter vastly at odds with common practice. (d) There is a better alternative. A regression of the variable at date t on the four most recent values as of date t - h achieves all the objectives sought by users of the HP filter with none of its drawbacks.

Subfields: Economics and Econometrics, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/cfb2dd686e61df3a8c6d172c0e9761f84669611b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

In Electrical and Electronic Engineering

Citations in the corpus

BUSINESS CYCLES, TREND ELIMINATION, AND THE HP FILTER (2020, methodology): “d (L)−1 = ( 1 + λ∆2∆∗2 )−1 diag [O2, In−4, O2] + ( 1 + λ∆2∆∗2 )−1 (17) × ( Ea [I2 + αλ (L)A (L)] −1E′ a + Eb [I2 + αλ (L)B (L)] −1E′ b ) ,”
Economic uncertainty before and during the COVID-19 pandemic (2020, methodology): “We detrend all variables using the method of Hamilton (2018), with p = 36 and h = 12.”
In Search of the Origins of Financial Fluctuations: The Inelastic Markets Hypothesis (2021, methodology): “This would make it possible to estimate the causal impact of prices on capital flows, and of capital flows on prices. then extract the cyclical component using the methodology developed in Hamilton (2018).”
Learning from History: Volatility and Financial Crises (2016, methodology): “This implies using a one-sided HP filter, con(5)Alternatively, one could adopt Markov switching models, for modeling the trend levels of volatility, along the lines of Hamilton and Susmel (1994). However, given our sample size, with Markov switching we are limited to most two regimes, and in additio”
Modeling fluctuations in the global demand for commodities (2018, neutral): “the identification of aggregate demand shocks (see, e.g., Diebold and Kilian 2001, Canova 2014; Hamilton 2018).”
From boom to bust: a typology of real commodity prices in the long run (2013, neutral): “…commodity prices have continued to rise, even in the face of the significant reversal in real commodity prices dating from 2011; (2) recent work by Hamilton (2017) strongly advises against the use of the HP filter in that it “produces series with spurious dynamic relations that have no basis in…”
Global Financial Cycles and Risk Premiums (2018, neutral): “As a robustness check, we also report results based on an alternative nonparametric detrending method recently suggested by Hamilton (2018).”
Corruption and economic growth: New empirical evidence (2019, neutral): “Our baseline model specifies = 4 , which is based on the recommendation of Hamilton (2018), who shows that using four lags of � removes unit roots in the time series in Equation (1).”
Coronavirus: Impact on Stock Prices and Growth Expectations (2020, neutral): “To illustrate the relation between the two series at the business-cycle frequency, we extract the cyclical component of real dividends and real GDP using the methodology developed in Hamilton (2018) zt = d0 + 11∑ j=8 dizt−j + ct, where zt corresponds to either log real dividends or log real GDP.”
Credit-Market Sentiment and the Business Cycle (2017, neutral): “Columns (3) and (4) then replicate columns (1) and (2), but instead of using five-year growth rates of the balance-sheet variables, they use “gaps,” where these gaps are defined as the deviation of the log-level of the variable from its trend, which is in turn estimated using the linear projection m”