Is the Hamilton regression filter really superior to Hodrick-Prescott detrending?
Identifikátory výsledku
Kód výsledku v IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11230%2F24%3A10490059" target="_blank" >RIV/00216208:11230/24:10490059 - isvavai.cz</a>
Výsledek na webu
<a href="https://verso.is.cuni.cz/pub/verso.fpl?fname=obd_publikace_handle&handle=aoT7TlhEYj" target="_blank" >https://verso.is.cuni.cz/pub/verso.fpl?fname=obd_publikace_handle&handle=aoT7TlhEYj</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1017/S136510052400018X" target="_blank" >10.1017/S136510052400018X</a>
Alternativní jazyky
Jazyk výsledku
angličtina
Název v původním jazyce
Is the Hamilton regression filter really superior to Hodrick-Prescott detrending?
Popis výsledku v původním jazyce
An article published in 2018 by J.D. Hamilton gained significant attention due to its provocative title, "Why you should never use the Hodrick-Prescott filter." Additionally, an alternative method for detrending, the Hamilton regression filter (HRF), was introduced. His work was frequently interpreted as a proposal to substitute the Hodrick-Prescott (HP) filter with HRF, therefore utilizing and understanding it similarly as HP detrending. This research disputes this perspective, particularly in relation to quarterly business cycle data on aggregate output. Focusing on economic fluctuations in the United States, this study generates a large amount of artificial data that follow a known pattern and include both a trend and cyclical component. The objective is to assess the effectiveness of a certain detrending approach in accurately identifying the real decomposition of the data. In addition to the standard HP smoothing parameter of $lambda = 1600$ , the study also examines values of $lambda <^>{star }$ from earlier research that are seven to twelve times greater. Based on three unique statistical measures of the discrepancy between the estimated and real trends, it is evident that both versions of HP significantly surpass those of HRF. Additionally, HP with $lambda <^>{star }$ consistently outperforms HP-1600.
Název v anglickém jazyce
Is the Hamilton regression filter really superior to Hodrick-Prescott detrending?
Popis výsledku anglicky
An article published in 2018 by J.D. Hamilton gained significant attention due to its provocative title, "Why you should never use the Hodrick-Prescott filter." Additionally, an alternative method for detrending, the Hamilton regression filter (HRF), was introduced. His work was frequently interpreted as a proposal to substitute the Hodrick-Prescott (HP) filter with HRF, therefore utilizing and understanding it similarly as HP detrending. This research disputes this perspective, particularly in relation to quarterly business cycle data on aggregate output. Focusing on economic fluctuations in the United States, this study generates a large amount of artificial data that follow a known pattern and include both a trend and cyclical component. The objective is to assess the effectiveness of a certain detrending approach in accurately identifying the real decomposition of the data. In addition to the standard HP smoothing parameter of $lambda = 1600$ , the study also examines values of $lambda <^>{star }$ from earlier research that are seven to twelve times greater. Based on three unique statistical measures of the discrepancy between the estimated and real trends, it is evident that both versions of HP significantly surpass those of HRF. Additionally, HP with $lambda <^>{star }$ consistently outperforms HP-1600.
Klasifikace
Druh
J<sub>imp</sub> - Článek v periodiku v databázi Web of Science
CEP obor
—
OECD FORD obor
50201 - Economic Theory
Návaznosti výsledku
Projekt
—
Návaznosti
I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace
Ostatní
Rok uplatnění
2024
Kód důvěrnosti údajů
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Údaje specifické pro druh výsledku
Název periodika
Macroeconomic Dynamics
ISSN
1365-1005
e-ISSN
1469-8056
Svazek periodika
29
Číslo periodika v rámci svazku
May 2024
Stát vydavatele periodika
US - Spojené státy americké
Počet stran výsledku
14
Strana od-do
1-14
Kód UT WoS článku
001214586800001
EID výsledku v databázi Scopus
2-s2.0-85192814681