Can we use Google Trends data to predict exchange rate movements?

What is it about?

Is it possible to use internet search query data to proxy some of the exchange rate fundamentals before the official data releases and use that information to forecast the monthly exchange rate returns accordingly? In this paper, I try to capture people’s perception of various macroeconomic variables by analyzing the internet search volume data via Google Trends. I collected 16 search volume data for each country in our sample and applied three different tests to evaluate the performance of the Google Trends-based predictions against the conventional structural model-based ex-post predictions. In out-of-sample forecasting of monthly returns on exchange rates, my findings indicate that the Google Trends search query data do a better job than the structural models in predicting the true direction of changes in nominal exchange rates. I observed that Google Trends-based forecasts are better at picking up the direction of the changes in the monthly nominal exchange rates after the Great Recession era (2008–2009).

Featured Image

Why is it important?

By using internet search data, we can get a timely description of the state of the economy long before the official data are released to the market participants. Google Trends data can be a useful alternative data source for making some real time macro-based predictions.

Perspectives