I heard it through the newswire: Developing an economic policy uncertainty index for Finland based on newswire articles

Loading...
Thumbnail Image

URL

Journal Title

Journal ISSN

Volume Title

School of Business | Master's thesis

Date

2024

Major/Subject

Mcode

Degree programme

Information and Service Management (ISM)

Language

en

Pages

90 + 22

Series

Abstract

In this thesis, I utilize word2vec embeddings and Latent Dirichlet Allocation to create a long-term Finnish economic policy uncertainty index inspired by Baker, Bloom and Davis (2016) using news wire articles published by STT be-tween 1992 and 2021. I show that the resulting index and its subcomponents can be used to effectively highlight periods of heightened uncertainty and their causes throughout the sample. Descriptive analysis of the indices highlights the 90s, early 00s and late 10s as exceptionally uncertain periods during the sample. I use recursive SVAR models to highlight that the responses to impulses in domestic EPU indices are noticeably smaller in magnitude compared to responses to shocks in alternative uncertainty indices such as the realized volatility of the OMXH or international uncertainty indices. Subindex level analysis shows that shocks to policy-related subindices are not associated with meaningful macroeconomic responses, but shocks to business-related topics are linked to statistically significant declines in investment and GDP and correlate highly with international uncertainty indices. My results suggest that Finnish policy uncertainty is primarily domestic, and it does not have a major effect on the Finnish economy, while economically relevant uncertainty has a significant international component and is associated with major macroeconomic effects.

Description

Thesis advisor

Malo, Pekka

Keywords

uncertainty, EPU, LDA, topic modelling, macroeconomics

Other note

Citation