"DIY Forecasting: Judgment, Models and Judgmental Model Selection"
presented in INFORMS2014 - San Francisco, Cardiff, Lancs, LUBS, MBS, AUEB, NTUA, Monash and soon in Queen’s Management School(Belfast) - 16/03/16 and Roma Tre - 25/05/16

Cluster: Contributed
Session Information: Wednesday Nov 12, 16:30 - 18:00
Title: Forecasting 2
Chair: Semco Jahanbin,Doctoral Researcher, University of Bath, Flat 8, Royston house, 5 duke street, Bath ba24ah, United Kingdom, This email address is being protected from spambots. You need JavaScript enabled to view it.

 

Title: DYI Forecasting: Judgment, Models and Judgmental Model Selection

Presenting Author: Konstantinos Nikolopoulos,Bangor Business School, Bangor University, College Road, Bangor LL57 2DG, United Kingdom, This email address is being protected from spambots. You need JavaScript enabled to view it.
Co-Author: Nikolaos Kourentzes,Lancaster University, The Management School, Lancaster University, Lancaster LA1 4YX, United Kingdom, This email address is being protected from spambots. You need JavaScript enabled to view it.
Fotios Petropoulos,Cardiff Business School, Cardiff University, Aberconway Building, Cardiff CF10 3EU, United Kingdom, This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract: In this paper we explore how judgment can be used to improve statistical model selection for forecasting. We investigate the performance of various judgmental model selection methodologies against the benchmark statistical one, based on information criteria. We evaluate the performance of experts in terms of selecting the best model and forecasting performance, identifying major improvements. We examine how to extend statistical model selection to incorporate additional insights from experts.


Title: Determining an Optimal Hierarchical Forecasting Model Based on the Characteristics of the Data Set

Presenting Author: Zlatana Nenova,University of Pittsburgh, 241 Mervis Hall, Pittsburgh, United States of America, This email address is being protected from spambots. You need JavaScript enabled to view it.
Co-Author: Jerrold May,Professor of Business Administration, Katz Graduate School of Business, 278A Mervis Hall, Pittsburgh PA 15260, United States of America, This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract: High-dimensional pyramidal databases are common in the supply chains of large manufacturing companies. Such organizations often forecast shipments and consumption patterns at different hierarchical levels. Determining the most appropriate forecast aggregation approach is often a very computationally intensive task. Using a large food-processing firm data, we built a model that requires only correlation metrics and produces an accurate prediction of the optimal forecasting approach.


Title: How Change of the Relative Importance of Product Attribute to Consumers can Influence Sales Forecast

Presenting Author: Semco Jahanbin,Doctoral Researcher, University of Bath, Flat 8, Royston house, 5 duke street, Bath ba24ah, United Kingdom, This email address is being protected from spambots. You need JavaScript enabled to view it.
Co-Author: Paul Goodwin,Professor, University of Bath, University of Bath, School of Management, Claverton down, Bath BA27AY, United Kingdom, This email address is being protected from spambots. You need JavaScript enabled to view it.
Sheik Meeran,Senior Lecturer(Associate Professor), University of Bath, University of Bath, School of Management, Claverton down, Bath BA27AY, United Kingdom, This email address is being protected from spambots. You need JavaScript enabled to view it.
Joao Quariguasi Frota Net,Senior Lecturer(Associate Professor), University of Bath, University of Bath, School of Management, Claverton down, Bath BA27AY, United Kingdom, This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract: Customer preferences are not stable, especially where a consumer needs to make a complex or unfamiliar decision. In this research, the instability of consumer preferences for different attributes for a purposive sample of electronics products will be examined and compared from different angles with the aim of finding its influence on choice based conjoint analysis as a new product sales forecasting method.