Forecasting with news sentiment: Evidence with UK newspapers

Dooruj Rambaccussing; Andrzej Kwiatkowski

doi:10.1016/j.ijforecast.2020.04.002

Forecasting with news sentiment: Evidence with UK newspapers

Dooruj Rambaccussing (Lead / Corresponding author), Andrzej Kwiatkowski

University of Dundee School of Business

Research output: Contribution to journal › Special issue › peer-review

13 Citations (Scopus)

39 Downloads (Pure)

Abstract

We investigate the performance of newspapers for forecasting inflation, output and unemployment in the United Kingdom. We concentrate on whether the economic policy content reported in popular printed media can improve on existing point forecasts. We find no evidence supporting improved nowcasts or short-term forecasts for inflation. The sentiment inferred from printed media, can however be useful for forecasting unemployment and output. Considerable improvements are also noted when using individual newspapers and keyword based indices.

Original language	English
Pages (from-to)	1501-1516
Number of pages	16
Journal	International Journal of Forecasting
Volume	36
Issue number	4
Early online date	18 May 2020
DOIs	https://doi.org/10.1016/j.ijforecast.2020.04.002
Publication status	Published - 1 Oct 2020

Keywords

Textual Econometrics
Economic Sentiments
Economic News
Inflation
Inflation Reports
Output growth
Newpapers
Textual econometrics
Economic news
Newspapers
Inflation reports
Economic sentiments

ASJC Scopus subject areas

Business and International Management

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1016/j.ijforecast.2020.04.002Licence: UK Government Non-Commercial Licence

Author Accepted Manuscript
© 2020. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/
Accepted author manuscript, 3.2 MBLicence: CC BY-NC-ND

Cite this

@article{6095ca1846dc4da385b764943251e436,

title = "Forecasting with news sentiment: Evidence with UK newspapers",

abstract = "We investigate the performance of newspapers for forecasting inflation, output and unemployment in the United Kingdom. We concentrate on whether the economic policy content reported in popular printed media can improve on existing point forecasts. We find no evidence supporting improved nowcasts or short-term forecasts for inflation. The sentiment inferred from printed media, can however be useful for forecasting unemployment and output. Considerable improvements are also noted when using individual newspapers and keyword based indices.",

keywords = "Textual Econometrics, Economic Sentiments, Economic News, Inflation, Inflation Reports, Output growth, Newpapers, Textual econometrics, Economic news, Newspapers, Inflation reports, Economic sentiments",

author = "Dooruj Rambaccussing and Andrzej Kwiatkowski",

year = "2020",

month = oct,

day = "1",

doi = "10.1016/j.ijforecast.2020.04.002",

language = "English",

volume = "36",

pages = "1501--1516",

journal = "International Journal of Forecasting",

issn = "0169-2070",

publisher = "Elsevier",

number = "4",

}

TY - JOUR

T1 - Forecasting with news sentiment

T2 - Evidence with UK newspapers

AU - Rambaccussing, Dooruj

AU - Kwiatkowski, Andrzej

PY - 2020/10/1

Y1 - 2020/10/1

N2 - We investigate the performance of newspapers for forecasting inflation, output and unemployment in the United Kingdom. We concentrate on whether the economic policy content reported in popular printed media can improve on existing point forecasts. We find no evidence supporting improved nowcasts or short-term forecasts for inflation. The sentiment inferred from printed media, can however be useful for forecasting unemployment and output. Considerable improvements are also noted when using individual newspapers and keyword based indices.

AB - We investigate the performance of newspapers for forecasting inflation, output and unemployment in the United Kingdom. We concentrate on whether the economic policy content reported in popular printed media can improve on existing point forecasts. We find no evidence supporting improved nowcasts or short-term forecasts for inflation. The sentiment inferred from printed media, can however be useful for forecasting unemployment and output. Considerable improvements are also noted when using individual newspapers and keyword based indices.

KW - Textual Econometrics

KW - Economic Sentiments

KW - Economic News

KW - Inflation

KW - Inflation Reports

KW - Output growth

KW - Newpapers

KW - Textual econometrics

KW - Economic news

KW - Newspapers

KW - Inflation reports

KW - Economic sentiments

UR - http://www.scopus.com/inward/record.url?scp=85084727976&partnerID=8YFLogxK

U2 - 10.1016/j.ijforecast.2020.04.002

DO - 10.1016/j.ijforecast.2020.04.002

M3 - Special issue

SN - 0169-2070

VL - 36

SP - 1501

EP - 1516

JO - International Journal of Forecasting

JF - International Journal of Forecasting

IS - 4

ER -

Forecasting with news sentiment: Evidence with UK newspapers

Abstract

Keywords

ASJC Scopus subject areas

UN SDGs

Access to Document

Other files and links

Fingerprint

Time Series of Media Implied Economic Sentiments (Dictionary)

Time Series of Media Implied Economic Sentiments (Linear Support Vector Machine)

Kwiatkowski, Andrzej

Cite this

Forecasting with news sentiment: Evidence with UK newspapers

Abstract

Keywords

ASJC Scopus subject areas

UN SDGs

Access to Document

Other files and links

Fingerprint

Datasets

Time Series of Media Implied Economic Sentiments (Dictionary)

Time Series of Media Implied Economic Sentiments (Linear Support Vector Machine)

Profiles

Kwiatkowski, Andrzej

Cite this