Led by: Professor Jules H. van Binsbergen, The Nippon Life Professor in Finance at the Wharton School

In this webinar, Professor van Binsbergen will discuss how he and other researchers used state-of-the-art machine learning methods to create an algorithm that predicts economic fundamentals such as GDP (both nationally and locally), consumption, and employment growth, even after controlling for commonly-used predictors; it also materially predicts monetary policy decisions, particularly during recessions. The measure relies on text of 200 million pages from 13,000 US local newspapers to construct a novel 170-year-long time series measure of economic sentiment at the country and state levels. This innovative application of machine learning techniques on a large historical data set has allowed for a better understanding of the role that sentiment has on economic activity as well as an improved predictive power for economic forecasts.

December 6, 2023