Work detail

Web search as an indicator of consumer confidence in the Visegrad Group

Author: Bc. Josef Matyáš
Year: 2018 - summer
Leaders: prof. PhDr. Ladislav Krištoufek Ph.D.
Consultants:
Work type: Bachelors
Language: English
Pages: 77
Awards and prizes:
Link: https://is.cuni.cz/webapps/zzp/detail/191642/
Abstract: This thesis examines the potential of utilizing web search information to nowcast consumer
condence indicators in the Visegrad Group countries. Data capturing search volume in all
available categories on Google search engine are analyzed and their potential to improve
forecasting of contemporaneous values of consumer condence indicators is assessed in
augmented autoregressive models. Information from categories with the most potential
is aggregated with utilization of principal component analysis and 3 sets of models are
examined in out-of-sample setting through measures of predictive accuracy and DieboldMariano
test. In-sample, models with Google data exhibited large improvements of t
in all countries. Predictive potential in simulation of real-time setting was conrmed in
Poland for categories related to automotive and consumer electronics. Limited potential
of category related to news in business was found in Hungary. In Czech Republic and
Slovakia, consistent nowcasting potential was not found for any category.

Partners

Deloitte

Sponsors

CRIF
McKinsey
Patria Finance