Editor-in-Chief Len Tashman's Preview of the Summer 2019 Issue of Foresight
This 54th edition of Foresight features an important new take on the role of judgment in the forecasting process. Normally, we see judgment applied to adjust a statistical forecast or supply a forecast directly. But here Fotios Petropoulos proposes a major role for judgment in the model-building process, providing cogent evidence for greater user involvement in selecting a forecasting model. We have entitled his article Judgmental Model Selection.
We follow the article with four Commentaries:
- Paul Goodwin suggests that the results “should provide reassurance to forecasters who choose which models to use rather than relying on the automatic selection of their software.”
- Nigel Harvey probes the basis of “algorithmic aversion.”
- Eric Stellwagen questions the practicality of judgmental model selection when there is a large variety of
models to choose from.
- Spyros Makridakis recommends extending the analysis to the time series in the M4 Competition.
A new book—Data Science for (the) Supply Chain Forecast—by Nicolas Vandeput offers a broad introduction to the application of machine-learning techniques to business forecasting. Reviewer Shaun Snapp believes the book effectively prompts the reader to begin performing “data science.”
Our Forecasting Methods Tutorial in this issue is an as “user friendly as possible” description of the State Space Framework for forecasting, which has long been in the statistician’s tool kit but inaccessible to practitioners. In State Space Modeling for Practitioners, Diego Pedregal provides a comprehensive overview of the software available and gives recommendations on what may best serve different types of users.
Our sister publication The International Journal of Forecasting presented a feature section earlier this year on prediction markets and their efficacy for business forecasting. Here, Thomas Wolfram summarizes the findings reported in the IJF but expands upon the Benefits and Challenges of Corporate Prediction Markets. He suggests that Big Data may be a more productive avenue for business forecasters to pursue.
Our section on Forecasting Practice includes two articles of a complementary nature. First, Chris Gray does a soul search into Why Is It So Hard to Hold Anyone Accountable for the Sales Forecast? This article offers “reality therapy” to the forecasting methodologist, uncovering the multitude of voices (and excuses) that may be heard before a final forecast is reached.
Picking up on the final step of the forecasting process, Alec Finney argues that the forecaster’s responsibility is much more than presentation of the numbers in Communicating the Forecast: Providing Decision Makers with Insights.
The 2019 Foresight Practitioner Conference
Mark your calendar for November 13-14 for this year’s Foresight Practitioner Conference, where in beautiful Chapel Hill, North Carolina, a distinguished group of speakers will offer their insights into Artificial Intelligence: The Hype and the Promise for Forecasting and Planning. See details on the conference program and registration on the inside cover of this issue. I’ll be cohosting the conference with Foresight Forecasting Practice Editor Mike Gilliland of SAS. Hope to see you there!
The Foresight Advisory Board
Joining the FAB is Niels van Hove, noted expert on Sales and Operations Planning and author of several articles and commentaries on S&OP for Foresight. Niels’ fascinating breadth of interests comes through in his Forecaster in the Field interview featured in our Summer 2016 issue.
And rejoining the FAB is legacy supporter Joe McConnell, principal of McConnell Chase Software Works. Joe has spent the last five years bringing to fruition the prodigious FD-7 package for forecasting, inventory optimization, and demand and supply planning.