Testing Investment Forecast Efficiency with Forecasting Narratives

Alexander Foltas

Journal of Economics and Statistics (Jahrbuecher fuer Nationaloekonomie und Statistik), 2022, vol. 242, issue 2, 191-222

Abstract: I analyze the narratives that accompany business cycle forecasting reports of three German institutes using topic models. To this end, I gather multiple similar topics into different economic subject categories, allowing me to map shifting prioritizations within and between these subjects. Subsequently, I examine whether forecasting narratives contain additional information not captured by traditional indicators and include them in a random forest-based investment-forecast efficiency analysis. I find multiple correlations between narratives and forecast errors and conclude that forecasters inefficiently incorporate qualitative information in these cases. I raise the idea that further investigations with more precise identification of forecasting narratives could improve qualitative information processing or lead to scientific guidelines for forecast adjustments.

Keywords: forecast efficiency; investment; random forrest; topic modeling; forecasting narratives; forecast adjustments (search for similar items in EconPapers)
JEL-codes: C53 E22 E27 (search for similar items in EconPapers)
Date: 2022
References: Add references at CitEc
Citations:

Downloads: (external link)
https://doi.org/10.1515/jbnst-2020-0027 (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:jns:jbstat:v:242:y:2022:i:2:p:191-222:n:3

DOI: 10.1515/jbnst-2020-0027

Access Statistics for this article

Journal of Economics and Statistics (Jahrbuecher fuer Nationaloekonomie und Statistik) is currently edited by Peter Winker

More articles in Journal of Economics and Statistics (Jahrbuecher fuer Nationaloekonomie und Statistik) from De Gruyter
Bibliographic data for series maintained by Peter Golla ().