Can Google Search Data Help Predict Macroeconomic Series?

Abstract

We use Google search data with the aim of predicting unemployment, CPI and consumer confidence for the US, UK, Canada, Germany and Japan. Google search queries have previously proven valuable in predicting macroeconomic variables in an in-sample context. To our knowledge, the more challenging question of whether such data have out-of-sample predictive value has not yet been satisfactorily answered. We focus on out-of-sample nowcasting, and extend the Bayesian Structural Time Series model using the Hamiltonian sampler for variable selection. We find that the search data retain their value in an outof-sample predictive context for unemployment, but not for CPI and consumer confidence. It may be that online search behaviour is a relatively reliable gauge of an individual's personal situation (employment status), but less reliable when it comes to variables that are unknown to the individual (CPI) or too general to be linked to specific search terms (consumer confidence).

Fear of hazards in commodity futures markets

“…Using Google searches by gold price and oil price as keywords, Vozlyublennaia (2014) finds that more attention decreases predictability (the ability of current/past returns to convey information about future returns) and thus argues that pricing efficiency increases. equity returns (Da et al, 2011(Da et al, , 2015Ben-Rephael et al, 2017;Dzielinski et al, 2018), sovereign credit spreads (Dergiades et al, 2015), and macroeconomic variables such as unemployment (D'Amuri and Marcucci, 2017;Niesert et al, 2019) inter alia.…”

Can Google Search Data Improve the Unemployment Rate Forecasting Model? An Empirical Analysis for Turkey

“…GT Data is used in studies in a lot of different fields, from flu outbreaks in the field of medicine (Ginsberg, Mohebbi, Patel, Brammer, Smolinski & Brilliant, 2008), exchange rate movements in economics and finance (Goddard, Kita, & Wang, 2015;Smith, 2012), stock markets (Da, Engelberg & Gao, 2011;Aouadi, Arouri & Teulon, 2013;Mondria, Wu & Zhang, 2010;Hamid & Heiden, 2015;Bui & Nguyen, 2019;Huang, Rojas & Convery, 2020), housing prices (Beracha & Wintoki, 2013), private consumption spending (Kholodilin, Podstawski & Siliverstovs, 2010), tourist movements (Bangwayo-Skeete & Skeete, 2015), prices of gold and oil (Han, Lv, & Yin, 2017;Jain & Biswal, 2019), and uncertainty (Donadelli & Gerotto, 2019;Castelnuovo & Tran, 2017), to consumer confidence (Niesert, Oorschot, Veldhuisen, Brons & Lange, 2019). With the forecasting area of macroeconomic indicators expanding with GT Data, several studies on unemployment forecasting are of significance in the literature of analysis that uses Google data as the flagbearer.…”

How to improve oil consumption forecast using google trends from online big data?: the structured regularization methods for large vector autoregressive model

“…Online big data, google trend data is the search interest relative to the highest interest for the specific search term, that people search on google, in the specific region for one month. Many studies demonstrated that the google trend data improves forecast performance not only for oil markets but also for other economic or financial data: Guo and Ji, 2013;Fantazzi and Fomicher, 2014;Li et al (2015) for oil market; Carriere-Swallow and Labbe (2013) for now casting in emerging market; Bulut (2018) for exchange rate; Niesert et al (2020) for unemployment and many others.…”