Bayesian Herd Detection for Dynamic Data

59 Pages Posted: 18 Mar 2020 Last revised: 1 Mar 2023

See all articles by Jussi Keppo

Jussi Keppo

National University of Singapore (NUS) - NUS Business School

Ville Satopää

INSEAD - Technology and Operations Management

Date Written: June 10, 2021

Abstract

This article analyzes multiple agents who forecast an underlying dynamic state based on streams of (partially overlapping) information. Each agent minimizes a convex combination of their forecasting error and deviation from the other agents' forecasts. As a result, the agents exhibit herding behavior - a bias that has been well-recognized in the economics and psychology literature. Our first contribution is a general framework for analyzing agents' forecasts under different levels of herding. The underlying state dynamics can be non-linear with seasonality, trends, shocks, or other time-series components. Our second contribution describes how models within our framework can be estimated from data. We apply our estimation procedure to surveys of equity price forecasts and find that the agents concentrate on average 37% of their efforts on making similar forecasts. However, there is substantial variation in the level of herding over time; even though herding falls substantially during the 2007-2008 financial crisis, it rises after the crisis.

Keywords: Bayesian statistics, Dynamic modeling, Gaussian process, Judgmental forecasting, Imperfect information game

Suggested Citation

Keppo, Jussi and Satopää, Ville, Bayesian Herd Detection for Dynamic Data (June 10, 2021). Available at SSRN: https://ssrn.com/abstract=3542129 or http://dx.doi.org/10.2139/ssrn.3542129

Jussi Keppo

National University of Singapore (NUS) - NUS Business School ( email )

Mochtar Riady Building
15 Kent Ridge Drive
Singapore, 119245
Singapore

HOME PAGE: http://https://www.jussikeppo.com

Ville Satopää (Contact Author)

INSEAD - Technology and Operations Management ( email )

Boulevard de Constance
77 305 Fontainebleau Cedex
France

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
385
Abstract Views
1,872
Rank
159,157
PlumX Metrics