Everything OpenForecast publishes is open and free: books, lecture notes, peer-reviewed papers, theory articles, and more than a decade of applied posts. This page is the map of the learning materials on this site — each entry notes who it is for, so you can find your door quickly.
Books and lecture notes
Forecasting and Analytics with ADAM
The monograph behind our methods: from the basics of forecasting up to ADAM, the Augmented Dynamic Adaptive Model, encompassing exponential smoothing, ARIMA, and regression with advanced features. This is the full methodology behind the smooth package, in the open.
Start here to understand exactly what our models do and why.
Statistics for Business Analytics
Lecture notes for a module on statistics, gradually evolving into a textbook. The foundations: distributions, estimation, hypothesis testing, and regression — explained for analysts rather than mathematicians.
Start here for the statistical grounding beneath the forecasting material.
Papers
Peer-reviewed research
The published work underneath the methods: state-space models for intermittent demand, demand classification, model selection and estimation. Every single method we used has a rigorous research with a thorough analysis behind it.
For researchers, and for practitioners who want the primary source rather than the summary.
Learn forecasting
Forecasting theory
Posts on what to forecast, how, and why: the thinking behind the methods rather than the mechanics of running them.
For practitioners and students who want to make better modelling decisions, not just execute recipes.
Intermittent demand
What makes demand intermittent, how it is properly classified, and why most of the standard toolkit — including the benchmarks and error measures — behaves badly on it. Spare parts and slow movers live here.
For anyone forecasting products that sell rarely and unpredictably, and for anyone who has been told their data is “intermittent and lumpy”.
Forecast evaluation
Which error measures to trust, which to avoid, and what each of them actually shows — a question most tools and courses never properly answer. Getting it wrong quietly might harm the decisions you make.
For anyone who has to judge whether a forecast is any good.
The blog
The largest resource on this site: more than a decade of articles on forecasting methods, intermittent demand, package releases, conference notes, and applied examples in R and Python.
The category and tag lists in the sidebar are the fastest way to find something specific.
Looking for the software itself? See Packages. For talks, slides, and where we present next, see Events.
And if you would rather have all of this applied to your business than read about it, let’s talk.
Prefer to write? Send us a message