The bias bias

Henry Brighton, Gerd Gigerenzer

Research output: Contribution to journalArticleScientificpeer-review

Abstract

In marketing and finance, surprisingly simple models sometimes predict more accurately than more complex, sophisticated models. Here, we address the question of when and why simple models succeed - or fail - by framing the forecasting problem in terms of the bias-variance dilemma. Controllable error in forecasting consists of two components, the "bias" and the "variance". We argue that the benefits of simplicity are often overlooked because of a pervasive "bias bias": the importance of the bias component of prediction error is inflated, and the variance component of prediction error, which reflects an oversensitivity of a model to different samples from the same population, is neglected. Using the study of cognitive heuristics, we discuss how to reduce variance by ignoring weights, attributes, and dependencies between attributes, and thus make better decisions. Bias and variance, we argue, offer a more insightful perspective on the benefits of simplicity than Occam''s razor.
Original languageEnglish
Pages (from-to)1772-1784
Number of pages13
JournalJournal of Business Research
Volume68
DOIs
Publication statusPublished - 1 Aug 2015
Externally publishedYes

Fingerprint

prediction
heuristics
finance
marketing
attribute
decision

Keywords

  • Bias bias
  • Bias-variance dilemma
  • Occam's razor
  • Out-of-sample prediction
  • Simple heuristics
  • Uncertainty

Cite this

Brighton, Henry ; Gigerenzer, Gerd. / The bias bias. In: Journal of Business Research. 2015 ; Vol. 68. pp. 1772-1784.
@article{2d2c5dae83c6486bbea123bf0c147c77,
title = "The bias bias",
abstract = "In marketing and finance, surprisingly simple models sometimes predict more accurately than more complex, sophisticated models. Here, we address the question of when and why simple models succeed - or fail - by framing the forecasting problem in terms of the bias-variance dilemma. Controllable error in forecasting consists of two components, the {"}bias{"} and the {"}variance{"}. We argue that the benefits of simplicity are often overlooked because of a pervasive {"}bias bias{"}: the importance of the bias component of prediction error is inflated, and the variance component of prediction error, which reflects an oversensitivity of a model to different samples from the same population, is neglected. Using the study of cognitive heuristics, we discuss how to reduce variance by ignoring weights, attributes, and dependencies between attributes, and thus make better decisions. Bias and variance, we argue, offer a more insightful perspective on the benefits of simplicity than Occam''s razor.",
keywords = "Bias bias, Bias-variance dilemma, Occam's razor, Out-of-sample prediction, Simple heuristics, Uncertainty",
author = "Henry Brighton and Gerd Gigerenzer",
year = "2015",
month = "8",
day = "1",
doi = "10.1016/j.jbusres.2015.01.061",
language = "English",
volume = "68",
pages = "1772--1784",
journal = "JBR: Journal of Business Research",
issn = "0148-2963",
publisher = "Elsevier Inc.",

}

The bias bias. / Brighton, Henry; Gigerenzer, Gerd.

In: Journal of Business Research, Vol. 68, 01.08.2015, p. 1772-1784.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - The bias bias

AU - Brighton, Henry

AU - Gigerenzer, Gerd

PY - 2015/8/1

Y1 - 2015/8/1

N2 - In marketing and finance, surprisingly simple models sometimes predict more accurately than more complex, sophisticated models. Here, we address the question of when and why simple models succeed - or fail - by framing the forecasting problem in terms of the bias-variance dilemma. Controllable error in forecasting consists of two components, the "bias" and the "variance". We argue that the benefits of simplicity are often overlooked because of a pervasive "bias bias": the importance of the bias component of prediction error is inflated, and the variance component of prediction error, which reflects an oversensitivity of a model to different samples from the same population, is neglected. Using the study of cognitive heuristics, we discuss how to reduce variance by ignoring weights, attributes, and dependencies between attributes, and thus make better decisions. Bias and variance, we argue, offer a more insightful perspective on the benefits of simplicity than Occam''s razor.

AB - In marketing and finance, surprisingly simple models sometimes predict more accurately than more complex, sophisticated models. Here, we address the question of when and why simple models succeed - or fail - by framing the forecasting problem in terms of the bias-variance dilemma. Controllable error in forecasting consists of two components, the "bias" and the "variance". We argue that the benefits of simplicity are often overlooked because of a pervasive "bias bias": the importance of the bias component of prediction error is inflated, and the variance component of prediction error, which reflects an oversensitivity of a model to different samples from the same population, is neglected. Using the study of cognitive heuristics, we discuss how to reduce variance by ignoring weights, attributes, and dependencies between attributes, and thus make better decisions. Bias and variance, we argue, offer a more insightful perspective on the benefits of simplicity than Occam''s razor.

KW - Bias bias

KW - Bias-variance dilemma

KW - Occam's razor

KW - Out-of-sample prediction

KW - Simple heuristics

KW - Uncertainty

U2 - 10.1016/j.jbusres.2015.01.061

DO - 10.1016/j.jbusres.2015.01.061

M3 - Article

VL - 68

SP - 1772

EP - 1784

JO - JBR: Journal of Business Research

JF - JBR: Journal of Business Research

SN - 0148-2963

ER -