Macroeconomic noise removal algorithm (MARINER)

    Marinko Skare Affiliation


Standard econometric filters fail to extract explicit trend component from macroeconomic data series. Isolated cycles provide no economic interpretation of the extracted component. Adding new data to the sample (filtering) period results in instability of extracted components. This study proposes a new econometric filtering technique (MARINER) able to overcome known shortcomings in standard econometrics filters such as Hodrick and Prescott (1997), Baxter and King (1999), Christiano and Fitzgerald (2003). MARINER provides a practical tool for policy makers dealing with business cycles. It also provides economic interpretation (new theory) on causes and sources of business cycles elaborating on theories developed by Phillips (1962) and Škare (2010). MARINER decomposes GDP macroeconomic data series in trend (long term) and cycles (medium term) components using three year moving average recursive filtering method. Extracted cycles are defined as deviations from equilibrium GDP path (minimized output gap) caused by poor synchronization between monetary and fiscal policy. MARINER bridge the gap in the literature on measuring and causes of business cycles. MARINER can purpose as foundation for building a new, primer econometric filtering methods.

Keyword : business cycles, econometric filters, band-pass, moving average, MARINER, Golden triangle

How to Cite
Skare, M. (2017). Macroeconomic noise removal algorithm (MARINER). Technological and Economic Development of Economy, 23(3), 549-565.
Published in Issue
May 8, 2017
Abstract Views
PDF Downloads