Nowcasting Prices Using Google Trends: An Application to Central America

Abstract

The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent.

Google Trends and the forecasting performance of exchange rate models

“…Google provides weekly search volume data by analyzing a random portion of Google searches and creating an index between 0 and 100 by computing how many searches have been done for the terms entered into the Google search engine, relative to the total number of searches done on Google over the same time for the same geographic region. Since Google only looks at a random fraction of all searches in the specified geographic region, the sampling noise problem does occur when a practitioner collects data for the same search phrase at different days and, as shown in Seabold and Coppola (), at times the noise in the data can be significant. We tried to smooth out the data noise by collecting data on different days and by taking the average of the historical data: We first collect the data on 12 different days during the month of February 2016, then we took the averages.…”

“…Retrieved on 25 March 2015 from https://www.netmarketshare.com/ search-engine-market-share.aspx?qprid=4&qpcustomd=0 3 WhileChoi and Varian (2012) found modest improvements in the prediction accuracy of US unemployment benefits,Tuhkuri (2016) found limited improvements with extended data coverage Askitas and Zimmermann (2009),. on the other hand, found a strong correlation between the Google search activity index and unemployment rates for the German labor market.4 Seabold and Coppola (2015) forecasted prices in Central America with Google search query data. In a related paper,Vosen and Schmidt (2011) showed that the forecasting performance of the Google Trends indicator is as good as those provided by the University of Michigan Consumer Sentiment Index and the Conference Board Consumer Confidence Index.…”

The Forecasting Power of Internet Search Queries in the Brazilian Financial Market

“…Li, Shang, Wang and Ma (2015) employ a MIDAS (Mixed Data Sampling) model to predict Chinese inflation using data combining different search query terms. Following this line of studies, the work of Seabold and Coppola (2015) investigated the possibility of using search query volume to predict food prices and consumer goods price series in Central America. The authors found significant results for the markets in Costa Rica, El Salvador and Honduras.…”

Can One Improve Now-Casts of Crop Prices in Africa? Google Can.

“…These studies, however, share one aspect: the use of GSQ data in the context of industrialized countries, where high Internet-adoption rates prevail. Two notable exemptions are Carrière-Swallow and Labbé (2013), who use GSQ data to now-cast automobile sales in Chile as well as Seabold and Coppola (2015), who nowcast consumer price indices and staple food prices in Costa Rica, El Salvador and Honduras.…”