Direct vs bottom–up approach when forecasting GDP: Reconciling literature results with institutional practice

Esteves, Paulo Soares

doi:10.1016/j.econmod.2013.04.020

Cited by 11 publications

(6 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hence, these results seem to support the view that it is hard to improve significantly over a simple autoregressive model for public consumption (see also Esteves, 2013). One should mention that quarterly public consumption in Portugal presents a relatively smooth profile as it results typically from the quarterly distribution of annual figures.…”

Section: Out-of-sample Period 2002q1-2015q4 2002q1-2007q4 2008q1-2015supporting

confidence: 74%

“…In particular, the accuracy of forecasting directly the aggregate is compared with the performance obtained via a bottom-up approach. Previous empirical work along this line includes Fair and Shiller (1990) for the United States GNP, Zellner and Tobias (2000) for GDP growth in industrialized countries, Marcellino et al (2003) for several euro area aggregates, Hubrich (2005) and Duarte and Rua (2007) for inflation in the euro area and Portugal, Esteves (2013) for euro area GDP, among others.…”

mentioning

confidence: 99%

See 1 more Smart Citation

A bottom-up approach for forecasting GDP in a data-rich environment

Dias

Pinheiro

Rua

2017

Applied Economics Letters

View full text Add to dashboard Cite

In an increasingly data-rich environment, the use of factor models for forecasting purposes has gained prominence in the literature and among practitioners. In this article, we extend the work of Dias, Pinheiro and Rua (2015) by assessing the forecasting behaviour of factor models to predict several GDP components and investigate the performance of a bottomup approach to forecast Portuguese GDP growth. We find supporting evidence of the usefulness of factor models and noteworthy forecasting gains when conducting a bottomapproach drawing on the main aggregates of GDP.

show abstract

Section: Out-of-sample Period 2002q1-2015q4 2002q1-2007q4 2008q1-2015supporting

confidence: 74%

mentioning

confidence: 99%

A bottom-up approach for forecasting GDP in a data-rich environment

Dias

Pinheiro

Rua

2017

Applied Economics Letters

View full text Add to dashboard Cite

show abstract

“…Our paper is closely related to Esteves (2013), Higgins (2014), Foroni and Marcellino (2014), Pareja et al (2020), Cobb (2020) and Giovannelli et al (2020). Pareja et al (2020) specify dynamic factor models for GDP and several expenditure components for the purpose of forecasting.…”

Section: Introductionmentioning

confidence: 91%

Macroeconometric forecasting using a cluster of dynamic factor models

Glocker

Kaniovski

2021

Empir Econ

View full text Add to dashboard Cite

We propose a modelling approach involving a series of small-scale factor models. They are connected to each other within a cluster, whose linkages are derived from Granger-causality tests. GDP forecasts are established across the production, income and expenditure accounts within a disaggregated approach. This method merges the benefits of large-scale macroeconomic and small-scale factor models, rendering our Cluster of Dynamic Factor Models (CDFM) useful for model-consistent forecasting on a large scale. While the CDFM has a simple structure, its forecasts outperform those of a wide range of competing models and of professional forecasters. Moreover, the CDFM allows forecasters to introduce their own judgment and hence produce conditional forecasts.

show abstract

“…Our main findings make a strong point in favor of forecasting an aggregate variable such as real GDP on a disaggregate component level instead of directly forecasting the aggregate variable itself. Esteves (2013) argues, that forecasters in policy institutions generally tend to opt for more disaggregate approaches owing to the fact that for communication purposes, they need a forecast "composition story" behind the GDP forecast. From a pure forecast accuracy perspective, it is not clear whether disaggregate forecasting approaches are superior to direct, aggregate approaches.…”

Section: Introductionmentioning

confidence: 99%

Short-Term Forecasting Economic Activity in Germany: A Supply and Demand Side System of Bridge Equations

Pinkwart¹

2018

SSRN Journal

View full text Add to dashboard Cite

Research QuestionAn accurate assessment of current economic conditions and the short-term outlook is an important input for policy makers and institutions such as central banks. Quantitative forecasts play an important role for this purpose. In practice, it is not only the expected growth rate of the aggregate real gross domestic product (GDP) which matters, but also the corresponding composition, i.e. the drivers of the economic expansion on the sectoral level or in terms of demand impulses. ContributionAddressing this need, we propose a comprehensive disaggregate approach for short-term forecasting economic activity in Germany. We explicitly take into account the way GDP is computed in the national accounts statistics, i.e. on the one hand by the supply or production side and on the other hand by the demand side. The GDP figures calculated by the two sides usually yield different results and the official GDP release is somewhere in between. We make use of this statistical procedure by separately modeling the two sides of GDP in a system of so-called bridge equations at the most disaggregate level available and combining the resulting two aggregate GDP forecasts. ResultsComparing several specification schemes in an out-of-sample forecast evaluation setup, we are able to find accurate forecasts for most of the underlying GDP components. We then show first, that both approaches already yield aggregate forecasts which are better than a simple benchmark forecast for forecast horizons of up to 28 weeks and second, that combining the production side and the demand side projections substantially improves the forecast performance, in particular for the shorter forecast horizons. Nichttechnische ZusammenfassungFragestellung Für wirtschaftspolitische Entscheidungsträger und Institutionen wie Zentralbanken ist eine genaue Einschätzung der aktuellen konjunkturellen Lage und der kurzfristigen Perspektiven eine wichtige Entscheidungsgrundlage. Dabei spielen quantitative Prognosen eine wichtige Rolle. In der Praxis steht nicht nur die erwartete Wachstumsrate des aggregierten realen Bruttoinlandsprodukts (BIP) im Fokus, sondern auch die dazugehörige Komposition und somit die sektoralen Antriebskräfte sowie die Nachfrageimpulse hinter dem Wirtschaftswachstum. BeitragUm diesen Bedarf zu bedienen, wird ein umfassender disaggregierter Ansatz für die Kurzfristprognose der gesamtwirtschaftlichen Aktivität in Deutschland vorgestellt. Dabei wird explizit berücksichtigt, dass das BIP in den Volkswirtschaftlichen Gesamtrechnungen einerseits über die Entstehungsseite und andererseits über die Verwendungsseite berechnet wird. Beide Rechenansätze liefern üblicherweise unterschiedliche Ergebnisse, und die offiziellen BIP-Zahlen liegen dazwischen. Dieser statistischen Vorgehensweise wird Rechnung getragen, indem die Entstehungs-und die Verwendungsseite des BIP getrennt voneinander in einem System von sogenannten Brückengleichungen auf der tiefsten verfügbaren Disaggregationsstufe modelliert und die beiden sich daraus ergebenden aggregierten B...

show abstract

Direct vs bottom–up approach when forecasting GDP: Reconciling literature results with institutional practice

Cited by 11 publications

References 16 publications

A bottom-up approach for forecasting GDP in a data-rich environment

A bottom-up approach for forecasting GDP in a data-rich environment

Macroeconometric forecasting using a cluster of dynamic factor models

Short-Term Forecasting Economic Activity in Germany: A Supply and Demand Side System of Bridge Equations

Contact Info

Product

Resources

About