Ensembles of Crowds and Computers: Experiments in Forecasting

46 Pages Posted: 14 Oct 2015 Last revised: 14 Dec 2015

See all articles by Germán G. Creamer

Germán G. Creamer

Stevens Institute of Technology, School of Business; Columbia University - Department of Computer Science

Yong Ren

Stevens Institute of Technology - School of Business

Yasuaki Sakamoto

AXA Direct Japan

Jeffrey V. Nickerson

Stevens Institute of Technology - School of Business

Date Written: June 25, 2015

Abstract

This paper explores the power of news sentiment to predict financial returns, in particular the returns of a set of European stocks. Building on past decision support work going back to the Delphi method this paper describes a text analysis expert weighting algorithm that aggregates the responses of both humans and algorithms by dynamically selecting the best response according to previous performance. The proposed system is tested through an experiment in which ensembles of experts, crowds and machines analyzed Thomson Reuters news stories and predicted the returns of the relevant stocks mentioned right after the stories appeared. The expert weighting algorithm was better than or as good as the best algorithm or human in most cases. The capacity of the algorithm to dynamically select best answers from humans and machines results in an evolving collective intelligence: the final decision is an aggregation of the best automated individual answers, some of these come from machines, and some from humans. Additionally, this paper shows that the groups of humans, algorithms, and expert weighting algorithms have associated with them particular news topics that these groups are good at making predictions from.

Keywords: Human machine ensembles, forecasting, Delphi method, crowdsourcing, machine learning

JEL Classification: C53, C63, G12, G14, F30, C9

Suggested Citation

Creamer, Germán G. and Ren, Yong and Sakamoto, Yasuaki and Nickerson, Jeffrey V., Ensembles of Crowds and Computers: Experiments in Forecasting (June 25, 2015). Stevens Institute of Technology School of Business Research Paper No. 2015–54. Available at SSRN: https://ssrn.com/abstract=2673129 or http://dx.doi.org/10.2139/ssrn.2673129

Germán G. Creamer (Contact Author)

Stevens Institute of Technology, School of Business ( email )

1 Castle Point on Hudson
Hoboken, NJ 07030
United States
2012168986 (Phone)

HOME PAGE: http://www.creamer-co.com

Columbia University - Department of Computer Science ( email )

New York, NY 10027
United States

Yong Ren

Stevens Institute of Technology - School of Business ( email )

Hoboken, NJ 07030
United States

Yasuaki Sakamoto

AXA Direct Japan ( email )

Japan

Jeffrey V. Nickerson

Stevens Institute of Technology - School of Business ( email )

Hoboken, NJ 07030
United States

Register to save articles to
your library

Register

Paper statistics

Downloads
182
Abstract Views
1,084
rank
170,796
PlumX Metrics