<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>The aim of this paper is to investigate interdependencies between the WIG20 index and economic policy uncertainty (EPU) related keywords quantified by a Google Trends search index. Tests for two periods from January 2015 till December 2019 and from June 2016 till May 2021 have been performed. This allowed the period of relative stability from the time of economic shock caused by the COVID-19 pandemics followed by various restrictions imposed by the governments to be distinguished.</p>
<p>A bivariate VAR model to selected search terms and the value of the WIG20 index was applied. After using AIC to establish the optimal number of lags the Granger causality test was performed. The increased empirical relationship has been confirmed between twelve EPU related terms and changes in the WIG20 index in the second period versus six terms for the pre-COVID period. It was also found that in the post-COVID period the intensity of reverse relations increased.</p>
</abstract>

The aim of this paper is to investigate interdependencies between the WIG20 index and economic policy uncertainty (EPU) related keywords quantified by a Google Trends search index. Tests for two periods from January 2015 till December 2019 and from June 2016 till May 2021 have been performed. This allowed the period of relative stability from the time of economic shock caused by the COVID-19 pandemics followed by various restrictions imposed by the governments to be distinguished.
A bivariate VAR model to selected search terms and the value of the WIG20 index was applied. After using AIC to establish the optimal number of lags the Granger causality test was performed. The increased empirical relationship has been confirmed between twelve EPU related terms and changes in the WIG20 index in the second period versus six terms for the pre-COVID period. It was also found that in the post-COVID period the intensity of reverse relations increased.

How Google Trends can improve market predictions— the case of the Warsaw Stock Exchange

Department of Operations Research and Mathematical Economics

Poznań University of Economics and Business

Economics and Business Review

This work is licensed under the Creative Commons Attribution 4.0 International License.

The aim of this paper is to investigate interdependencies between the WIG20 index and economic policy uncertainty (EPU) related keywords quantified by a Google...

How Google Trends can improve market predictions— the case of the Warsaw Stock Exchange

Published Online: Jul 18, 2022

Page range: 7 - 28

Received: Nov 23, 2021

Accepted: Jun 25, 2022

DOI: https://doi.org/10.18559/ebr.2022.2.2

Keywords
Granger causality, Warsaw Stock Exchange, economic policy uncertainty, WIG20, Google Trends, predictions

© 2022 Paweł Kropiński et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

How Google Trends can improve market predictions— the case of the Warsaw Stock Exchange

Published Online: Jul 18, 2022

Page range: 7 - 28

Received: Nov 23, 2021

Accepted: Jun 25, 2022

DOI: https://doi.org/10.18559/ebr.2022.2.2

KeywordsGranger causality, Warsaw Stock Exchange, economic policy uncertainty, WIG20, Google Trends, predictions

© 2022 Paweł Kropiński et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Keywords
Granger causality, Warsaw Stock Exchange, economic policy uncertainty, WIG20, Google Trends, predictions