An Artificial Trend Index for Private Consumption Using Google Trends

In recent years, the use of databases that analyze trends, sentiments, or news to make economic projections or create indicators has gained significant popularity, particularly with the Google Trends platform. This article explores the potential of Google search data to develop a new index, the Artificial Consumption Trend Indicator (ACTI). The ACTI is a real-time leading indicator of private consumption, created using machine learning techniques to capture behavioral patterns from online search queries. Unlike traditional indicators, which are often delayed and based on historical survey or transaction data, the ACTI provides an up-to-date and forward-looking measure of consumer behavior. We focus on one of the key components of economic activity, private consumption (64% of GDP in Peru). By selecting and estimating categorized variables, our analysis demonstrates that Google data can identify patterns to generate a leading indicator and improve the accuracy of forecasts. Finally, the results show that Google’s Food and Tourism categories significantly reduce projection errors, highlighting the importance of using this information in a segmented manner to improve macroeconomic forecasts.