Online Program Home
My Program

Abstract Details

Activity Number: 274 - Macroeconomic Forecasting and Policy in Data Rich Digital Age Environments
Type: Invited
Date/Time: Tuesday, July 30, 2019 : 8:30 AM to 10:20 AM
Sponsor: Business and Economic Statistics Section
Abstract #300382
Title: Google Trends and the Macroeconomy: a Bayesian Mixed Frequency Approach
Author(s): Arnab Bhattacharjee* and David Kohns
Companies: Heriot-Watt University and Heriot-Watt University

This paper investigates the use of Google Trends for structural macroeconomic forecasting. An issue when dealing with Google Trends for forecasting is that of mixed frequency and high dimensionality. We provide structural mixed frequency frameworks as well as econometric frameworks which are able to deal with these issues. We synthesise the presented methods into one estimator, the Bayesian Structural Time Series model augmented MIDAS (BSTS-U-MIDAS), which is able to address both issues and uncover the economic relationship of high frequency variables to the low frequency target variable. We find that the estimator beats all other mixed frequency estimators and that Google Trends increase forecasting fit. This provides promising results for structural modeling using Google Trends.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program