We begin this issue with a tribute to our august – and newly retired – colleague, Roy Batchelor. From his “Primer on Neural Networks” going back to the second issue of Foresight (2005), Roy has composed no fewer than 18 articles, reviews, and commentaries for this journal. Memorable takeaways from these have been framed into our own version of a “Batchelor Party.” For this issue, we asked Roy to write a retrospective on his career in forecasting and finance, and you’ll see it here under the title “Konfessions of a Kibitzer.” Best wishes in retirement, Roy!
Our feature section addresses the challenge that often follows selection of a forecasting method and generation/evaluation of the forecasts: that of ensuring adoption of the forecasts by decision makers. Here, the data-science team at Target – Mahdi R. Yousefi, Stacey Faulkenberg Larsen, and Subramanian Iyer – presents an update on the company’s efforts to achieve buy-in to the DFE forecasts: “The Demand Forecasting Project at Target: Improving Collaboration and Adoption.”
In their article “Making Forecasts More Trustworthy,” Simon Spavound and Nikos Kourentzes describe four attributes that promote trust and argue for the wider exposure of data-science students to the practical realities of forecasting. A Commentary on their article from Paul Goodwin, Sinan Gönül, and Dilek Önkal embellishes the key attributes of intelligibility, alignment, reliability, and stability, and how these can be demonstrated and achieved.
Paul Goodwin’s Hot New Research column explains another approach to forecasting seasonal time series: “Subsampling Seasonal Series – A Simple Approach to Forecasting Complex Patterns.” Proposed by Xixi Li, Fotios Petropoulos, and Yanfei Kang, the method creates multiple forecasts by taking different subsamples of the time series, extrapolating each subsample and averaging the resulting forecasts.
Ira Sohn, Foresight’s Editor for Long-Range Forecasting, gives us two contributions to this issue. First, there’s his book review of the Atlas of Forecasts – a comprehensive and visually sumptuous distillation of global modeling efforts from their inception several thousand years ago to the present. Second, Ira ushered in the article from Keith Wiebe, Timothy Sulser, and Nicola Cenacchi of the International Food Policy Research Institute (IFPRI) on “Long-Term Projections of Food Production and Demand” and oversaw its development. Models created by IFPRI and other global teams have enabled study of long-term drivers of food systems, such as changes in population, income, technology, land, water, and climate.
To receive this issue, start your membership to the IIF today! Join Now