Skip to contentScholarGate
LibraryBookshelfDeskReview StudioAssistant
Sign in
On this page
IntuitionHow it worksWhen to use itStrengths & limitationsCommon pitfallsApplicationsFrequently asked🔒 Read the full methodSourcesRelated methods
Cite this pageSpotted an issue on this page? Report or suggest a fix →
Home›Econometrics›Hodrick-Prescott Filter: Trend-Cycle Decomposition for Macroeconomic Time Series
Process / pipelineTrend & seasonality

Hodrick-Prescott Filter: Trend-Cycle Decomposition for Macroeconomic Time Series

Hodrick-Prescott Filter · Also known as: Hodrick-Prescott Filter, HP Decomposition, Trend-Cycle Filter, HP Filtresi

The Hodrick-Prescott (HP) filter is a penalized least-squares technique used in macroeconomics and empirical finance to decompose a time series into a smooth long-run trend component and a short-run cyclical component. Introduced by Hodrick and Prescott (1997) using postwar U.S. business cycle data, it has become one of the most widely applied filters in business cycle analysis, monetary policy research, and applied econometrics.

ScholarGate
  1. Process / pipeline
  2. v1
  3. 1 Sources
  4. PUBLISHED
Cite this page →
Tools & resources
Download slides
Learn & explore

Read the full method

Members only

Sign in with a free account to read this section.

Sign in

Method map

The neighbourhood of related methods — select a node to explore.

HP Filter
BK FilterState Space ModelSTL Decomposition

When to use it

The HP filter is appropriate when the research objective is to decompose a macroeconomic or financial time series into trend and cycle, particularly for business cycle analysis, output gap measurement, or potential output estimation. It assumes the series is non-stationary with a slowly evolving trend and short-run cyclical deviations. Key assumptions are additivity of trend and cycle and a user-specified lambda consistent with the data frequency. It is less suitable for series with structural breaks, near-unit-root behavior, or end-point sensitivity concerns. Alternatives include the Baxter-King filter, STL decomposition, and Kalman-filter-based state-space models.

Strengths & limitations

Strengths
  • Simple closed-form computation requiring no iterative estimation; computationally fast even for long series.
  • Flexible across data frequencies through the lambda parameter (100/1600/14400 for annual/quarterly/monthly).
  • Produces interpretable, smooth trend series widely understood by policymakers and central bank economists.
  • Easily replicated and comparable across studies, facilitating cross-country and cross-period business cycle research.
Limitations
  • The choice of lambda is arbitrary and mechanical; results can vary substantially with different lambda values.
  • End-point problem: trend estimates near the beginning and end of the sample are unreliable due to asymmetric data availability.
  • Hamilton (2018) shows the HP filter can introduce spurious cycles into white-noise series, questioning its statistical validity.
  • Sensitive to unit roots and structural breaks; may produce misleading cycles when underlying data-generating process is misspecified.

Frequently asked

How do I choose the lambda parameter?

The conventional values recommended by Hodrick and Prescott are 100 for annual data, 1600 for quarterly data, and 14400 for monthly data. These were calibrated so that the ratio of business-cycle variance to trend variance matches empirical priors. For non-standard frequencies or specific research questions, some authors derive lambda via the formula lambda = 6.25 * (frequency ratio)^4, following Ravn and Uhlig (2002).

What is the end-point problem and how can I address it?

Because the HP filter is a two-sided filter, the trend estimates at the start and end of the sample depend on data that do not yet exist. This creates substantial revisions as new observations arrive. Common remedies include extending the sample with forecasted values before filtering, using one-sided variants of the filter for real-time applications, or switching to alternative filters such as the Hamilton (2018) regression-based approach that avoids the end-point issue.

Is the HP filter still acceptable given Hamilton's (2018) critique?

Hamilton (2018) demonstrated that the HP filter applied to unit-root processes induces spurious cycles even in white noise, and proposed a simple OLS regression alternative. Despite this critique, the HP filter remains widely used in policy institutions and comparative studies due to its computational simplicity and legacy. Researchers should acknowledge its limitations, complement results with alternative methods, and avoid strong structural interpretations of the extracted cycle component.

Sources

  1. Hodrick, R. J., & Prescott, E. C. (1997). Postwar U.S. business cycles: An empirical investigation. Journal of Money, Credit and Banking, 29(1), 1–16. DOI: 10.2307/2953682 ↗

How to cite this page

ScholarGate. (2026, June 2). Hodrick-Prescott Filter. ScholarGate. https://scholargate.app/en/econometrics/hp-filter

Related methods

BK FilterState Space ModelSTL Decomposition

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

  • BK FilterEconometrics↔ compare
  • State Space ModelEconometrics↔ compare
  • STL DecompositionEconometrics↔ compare
Compare side by side →

Referenced by

BK Filter

Similar methods

BK FilterMarkov-Switching ModelRegime-Switching ModelTime Series Kalman FilterTVP-FAVARStructural Time Series ModelState Space ModelSTL Decomposition

Related reference concepts

Prices, Business Fluctuations, and CyclesMacroeconomics and Monetary EconomicsTrend AnalysisMacroeconomicsEconometricsHidden Markov Models

Spotted an issue on this page? Report or suggest a fix →

ScholarGate — HP Filter (Hodrick-Prescott Filter). Retrieved 2026-07-21 from https://scholargate.app/en/econometrics/hp-filter · Dataset: https://doi.org/10.5281/zenodo.20539026
Quick facts
Originator
Robert Hodrick & Edward Prescott
Year
1997
Type
Penalized least-squares smoother
Subfamily
Trend & seasonality
Smoothing Parameter
λ (lambda); conventional values: 100 for annual, 1600 for quarterly, 14400 for monthly data
Related methods
BK FilterState Space ModelSTL Decomposition
ScholarGate

A content-first reference library for research methods — what each one is, how it works, and where it comes from.

Open data (CC-BY)

Explore

  • Library
  • Search the library…
  • Browse by field
  • Fields
  • Journey
  • Compare
  • Which method?

Reference

  • Subjects
  • Atlas
  • Glossary
  • Methodology
  • Philosophy

Your tools

  • Bookshelf
  • Desk
  • Chat

Company

  • About
  • Pricing
  • Contact
  • Suggest a method

Entries are compiled from published sources for reference. Verifying the accuracy and suitability of any information for your own use remains your responsibility.

© 2026 ScholarGate · A research-method reference library
  • Privacy
  • Cookies
  • Terms
  • Delete account