Assessing the Accuracy of Google Trends for Predicting Presidential Elections: The Case of Chile, 2006–2021

This article presents the results of reviewing predictive capacity Google Trends for national elections in Chile. The electoral between Michelle Bachelet and Sebastián Piñera 2006, Eduardo Frei 2010, Evelyn Matthei 2013, Alejandro Guillier 2017, Gabriel Boric José Antonio Kast 2021 were reviewed. time series analyzed organized on basis relative searches candidacies, assisted by R software, mainly with gtrendsR forecast libraries. With constructed, forecasts made using Auto Regressive Integrated Moving Average (ARIMA) technique to check weight one presidential option over other. ARIMA analyses performed 3 ways organizing data: linear series, transformed moving average, Hodrick–Prescott. indicate that method offers optimal ability.