Predictability analysis of the Pound's Brexit exchange rates based on Google Trends data

Area of Science:

• Economics
• Finance
• Data Science

Background:

• Online search traffic data, particularly Google Trends, is increasingly utilized for analyzing and predicting human behavior across various domains.
• The Sterling Pound's exchange rate predictability is a critical area of financial market research.
• The period studied (2015-2020) encompasses significant events like the UK referendum and Brexit.

Purpose of the Study:

• To explore the predictability of the Sterling Pound's exchange rates against the Euro and Dollar using Google Trends data.
• To analyze the relationship between Google search query data on Pound-related keywords and the currency's exchange rates.
• To investigate directional predictability from search trends to Pound exchange rate movements.

Main Methods:

• Utilized Google Trends data spanning five years (March 1, 2015, to February 29, 2020).
• Employed a quantile dependence method, specifically cross-quantilograms, to assess predictability.
• Tested for directional predictability across lags from zero to 30 weeks.

Main Results:

• Statistically significant quantile dependencies were identified between Google query data and Sterling Pound exchange rates.
• Evidence suggests a directional relationship from Google search trends to the Pound's exchange rate movements.
• The findings highlight the potential of search data to indicate reactions in economic variables.

Conclusions:

• Google Trends data offers valuable insights into the predictability of the Sterling Pound's exchange rates.
• The study demonstrates the utility of cross-quantilograms in uncovering complex economic relationships.
• Search query patterns can serve as an indicator for economic variable interactions and market movements.