Inflace: prediktivní síla dat Google Trends

Suchánek, Jan

This thesis explores the utility of Google Trends data in enhancing predictive power accuracy of ARIMA models for forecasting inflation in the Czech republic. The research was structured to address two core hypotheses: the rationality of inflation expectations as reflected in Google Trends search queries and the ability of the data to augment the predictive power of traditional inflation forecasting models. Our findings indicate that Google Trends data, when incorporated as an ex- ternal regressor in ARIMA models, significantly improve the model's predictive accuracy, especially in periods characterized by high inflation rates and eco- nomic turbulence. This provides evidence for the claim that Google Trends is able to effectively capture shifts in consumer sentiment and expectations. However, the study acknowledges limitations, including the specificity of the time domains analyzed and the exclusive focus on the Czech Republic. These factors may limit the generalizability of the results. In summary, this thesis contributes to the evolving field of economic fore- casting by demonstrating the value of integrating unconventional digital data sources like Google Trends into traditional econometric models. It opens av- enues for future research to explore the broader applicability of such data in...

host :: přihlásit Digitální repozitář
		Hledej		Nový záznam		Nápověda		O repozitáři