2015IMF Working PaperRequires access

A Simple Multivariate Filter for Estimating Potential Output

Patrick Blagrave, Roberto Garcia-Saltos, Douglas Laxton, Fan Zhang

Open publisher page 59 citations

Abstract

Estimates of potential output are an important ingredient of structured forecasting and policy analysis. Using information on consensus forecasts, this paper extends the multivariate filter developed by Benes and others (2010). Although the estimates in real time are more robust relative to those of naïve statistical filters, there is still significant uncertainty surrounding the estimates. The paper presents estimates for 16 countries and provides an example of how the filtered estimates at the end of the sample period can be improved with additional information.

About this research paper

What this paper is about

Estimates of potential output are an important ingredient of structured forecasting and policy analysis. Using information on consensus forecasts, this paper extends the multivariate filter developed by Benes and others (2010). Although the estimates in real time are more robust relative to those of naïve statistical filters, there is still significant uncertainty surrounding the estimates. The paper presents estimates for 16 countries and provides an example of how the filtered estimates at the end of the sample period can be improved with additional information.

Why it matters

OpenAlex reports 59 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Estimates of potential output are an important ingredient of structured forecasting and policy analysis. Using information on consensus forecasts, this paper extends the multivariate filter developed by Benes and others (2010). Although the estimates in real time are more robust relative to those of naïve statistical filters, there is still significant uncertainty surrounding the estimates. The paper presents estimates for 16 countries and provides an example of how the filtered estimates at the end of the sample period can be improved with additional information.

Key concepts: Multivariate statistics, Filter (signal processing), Econometrics, Simple (philosophy), Computer science, Hodrick–Prescott filter, Statistics, Sample (material)

Related papers

Back to paper searchBrowse research topicsOriginal source
A Simple Multivariate Filter for Estimating Potential Output — Research Paper | ScholarLens