Forecasts are everywhere.
Reliable decisions are not.
Advances in data, analytics, and artificial intelligence have made forecasting faster, more accessible, and increasingly automated. Models can be built, tuned, and compared in seconds. What once required expertise and time is now widely available.
Yet something fundamental has not changed:
The future remains unknown. Decisions are needed now—and no tool can change that.
This central challenge persists. Forecasts are plentiful, yet decisions remain difficult, inconsistent, and often misaligned with uncertainty.
The problem is not the lack of forecasts. It is the lack of systems that effectively translate forecasts into action.
For instructors and future forecasters, this creates both a challenge and an opportunity. The task is no longer simply to teach or learn analytical execution, but to develop judgment, structured reasoning, and decision design—capabilities that are essential in practice yet often underemphasized in traditional instruction.
This book addresses that gap by reorganizing forecasting education around three fundamental shifts:
These shifts reflect a broader reality: as prediction becomes easier, design and responsibility become more important.
They connect data and temporal structure, models and decision behavior, evaluation and trust, and ultimately, forecasts and decisions. The emphasis moves from technical execution to system design and decision capability.
Forecasting has traditionally been taught as a technical craft. Students learn methods, build models, and evaluate performance within a “toolbox” framework. This approach reflected a world in which analytical capability was scarce and inference was the primary barrier.
Today, that barrier has largely fallen. As analytical execution becomes easier, it becomes less differentiating. What distinguishes effective analysts is no longer their ability to produce forecasts, but their ability to use forecasts within decision contexts.
This book is built on a simple but consequential premise:
In an era of data abundance and advancing AI, the central challenge of forecasting is no longer producing optimal forecasts, but designing how imperfect forecasts guide decisions.
Forecasting operates within an uncontrollable reality: the future is unknown. Accuracy can only be evaluated after outcomes occur, yet decisions must be made before that point—and often under changing conditions.
In practice, forecasting is not just a modeling task. It is a decision system—one that shapes how organizations interpret uncertainty, allocate resources, and act.
This book reframes forecasting from a technical modeling exercise into a decision-support system that must be intentionally designed —an approach we call Forecast-by-Design.
Forecast-by-Design connects:
Alongside this reframing is a shift from training analytical techniques to developing decision capability.
Organizations have always required judgment, interpretation, and accountability. As analytical execution becomes easier, these capabilities become central:
Advances in data and AI do not replace these capabilities—they make them essential.
Rather than presenting forecasting as an inventory of techniques, this book uses selected methods to reveal how forecasting systems function in practice and to develop the ability to think, design, and apply them effectively.
In this book, AI is not treated as a tool for automation alone. It is integrated as a structured partner in learning and reasoning.
Each chapter includes a LearningLab organized around a three-mode framework:
Reinforce → Extend → Explore
This structure ensures that AI supports, rather than replaces, analytical thinking. It also allows foundational technical content to be developed, while classroom time focuses on structure, reasoning, and decision design.
A consistent principle applies throughout:
AI expands analytical range, but does not replace analytical responsibility.
Students are expected to verify claims, replicate reasoning, and recognize limitations—treating AI outputs as hypotheses, not conclusions.
This book is designed as a thinking system, not a collection of topics.
Students progress through:
Observe → Understand → Practice → Reason → Design → Decide → Integrate → Consolidate → Continue
This progression mirrors real-world analytical work and is implemented through structured chapter components, moving from analysis to decision design.
Forecast-by-Design is the central differentiator of this book. It is organized around a consistent analytical spine:
Structure → Behavior → Trust → Decision
Students learn not only how to build models, but how to:
The goal is not simply technical proficiency, but judgment under uncertainty.
This book does not replace traditional forecasting instruction. It extends it by introducing a new balance between AI-supported learning and decision-oriented capability.
It preserves quantitative rigor while reframing forecasting as a system of interpretation, judgment, and decision design in an AI-enabled environment.
The goal is not simply to teach forecasting.
It is to prepare students to recognize structure, design decisions, and act responsibly with forecasts in an age of AI.
Generative AI tools were used during the draft and revision process to assist with language polishing, readability improvement, stylistic refinement, and organizational editing. All conceptual framing, arguments, interpretations, and final editorial decisions remain the sole responsibility of the author.
