The sum and its parts: Judgmental hierarchical forecasting

Mirko Kremer, Enno Siemsen, Douglas J. Thomas

Research output: Contribution to journalArticle

22 Citations (Scopus)

Abstract

Firms require demand forecasts at different levels of aggregation to support a variety of resource allocation decisions. For example, a retailer needs store-level forecasts to manage inventory at the store, but also requires a regionally aggregated forecast for managing inventory at a distribution center. In generating an aggregate forecast, a firm can choose to make the forecast directly based on the aggregated data or indirectly by summing lower-level forecasts (i.e., bottom up). Our study investigates the relative performance of such hierarchical forecasting processes through a behavioral lens. We identify two judgment biases that affect the relative performance of direct and indirect forecasting approaches: a propensity for random judgment errors and a failure to benefit from the informational value that is embedded in the correlation structure between lowerlevel demands. Based on these biases, we characterize demand environments where one hierarchical process results in more accurate forecasts than the other.

Original languageEnglish (US)
Pages (from-to)2745-2764
Number of pages20
JournalManagement Science
Volume62
Issue number9
DOIs
StatePublished - Sep 2016

Fingerprint

Hierarchical forecasting
Relative performance
Retailers
Correlation structure
Bottom-up
Distribution center
Resource allocation
Demand forecast
Propensity

All Science Journal Classification (ASJC) codes

  • Strategy and Management
  • Management Science and Operations Research

Cite this

Kremer, Mirko ; Siemsen, Enno ; Thomas, Douglas J. / The sum and its parts : Judgmental hierarchical forecasting. In: Management Science. 2016 ; Vol. 62, No. 9. pp. 2745-2764.
@article{47b30a3f6723450c8970dcde73b3f974,
title = "The sum and its parts: Judgmental hierarchical forecasting",
abstract = "Firms require demand forecasts at different levels of aggregation to support a variety of resource allocation decisions. For example, a retailer needs store-level forecasts to manage inventory at the store, but also requires a regionally aggregated forecast for managing inventory at a distribution center. In generating an aggregate forecast, a firm can choose to make the forecast directly based on the aggregated data or indirectly by summing lower-level forecasts (i.e., bottom up). Our study investigates the relative performance of such hierarchical forecasting processes through a behavioral lens. We identify two judgment biases that affect the relative performance of direct and indirect forecasting approaches: a propensity for random judgment errors and a failure to benefit from the informational value that is embedded in the correlation structure between lowerlevel demands. Based on these biases, we characterize demand environments where one hierarchical process results in more accurate forecasts than the other.",
author = "Mirko Kremer and Enno Siemsen and Thomas, {Douglas J.}",
year = "2016",
month = "9",
doi = "10.1287/mnsc.2015.2259",
language = "English (US)",
volume = "62",
pages = "2745--2764",
journal = "Management Science",
issn = "0025-1909",
publisher = "INFORMS Inst.for Operations Res.and the Management Sciences",
number = "9",

}

Kremer, M, Siemsen, E & Thomas, DJ 2016, 'The sum and its parts: Judgmental hierarchical forecasting', Management Science, vol. 62, no. 9, pp. 2745-2764. https://doi.org/10.1287/mnsc.2015.2259

The sum and its parts : Judgmental hierarchical forecasting. / Kremer, Mirko; Siemsen, Enno; Thomas, Douglas J.

In: Management Science, Vol. 62, No. 9, 09.2016, p. 2745-2764.

Research output: Contribution to journalArticle

TY - JOUR

T1 - The sum and its parts

T2 - Judgmental hierarchical forecasting

AU - Kremer, Mirko

AU - Siemsen, Enno

AU - Thomas, Douglas J.

PY - 2016/9

Y1 - 2016/9

N2 - Firms require demand forecasts at different levels of aggregation to support a variety of resource allocation decisions. For example, a retailer needs store-level forecasts to manage inventory at the store, but also requires a regionally aggregated forecast for managing inventory at a distribution center. In generating an aggregate forecast, a firm can choose to make the forecast directly based on the aggregated data or indirectly by summing lower-level forecasts (i.e., bottom up). Our study investigates the relative performance of such hierarchical forecasting processes through a behavioral lens. We identify two judgment biases that affect the relative performance of direct and indirect forecasting approaches: a propensity for random judgment errors and a failure to benefit from the informational value that is embedded in the correlation structure between lowerlevel demands. Based on these biases, we characterize demand environments where one hierarchical process results in more accurate forecasts than the other.

AB - Firms require demand forecasts at different levels of aggregation to support a variety of resource allocation decisions. For example, a retailer needs store-level forecasts to manage inventory at the store, but also requires a regionally aggregated forecast for managing inventory at a distribution center. In generating an aggregate forecast, a firm can choose to make the forecast directly based on the aggregated data or indirectly by summing lower-level forecasts (i.e., bottom up). Our study investigates the relative performance of such hierarchical forecasting processes through a behavioral lens. We identify two judgment biases that affect the relative performance of direct and indirect forecasting approaches: a propensity for random judgment errors and a failure to benefit from the informational value that is embedded in the correlation structure between lowerlevel demands. Based on these biases, we characterize demand environments where one hierarchical process results in more accurate forecasts than the other.

UR - http://www.scopus.com/inward/record.url?scp=84988592835&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84988592835&partnerID=8YFLogxK

U2 - 10.1287/mnsc.2015.2259

DO - 10.1287/mnsc.2015.2259

M3 - Article

AN - SCOPUS:84988592835

VL - 62

SP - 2745

EP - 2764

JO - Management Science

JF - Management Science

SN - 0025-1909

IS - 9

ER -