Here is a curious idea: if we can somehow estimate the importance of trend/seasonal components for your data, you can use this in model building and forecasting. But how can we do this first step? Hans Levenbach has an answer with his simple EDA technique... openforecast.org/2026/05/18/h...

We continue our series of posts on the functions from the smooth package for Python/R. Today we will see how to enhance your exponential smoothing with explanatory variables. What? Yes, you heard me! Let’s dive in! openforecast.org/2026/05/05/s... #forecasting #datascience #machinelearning #python

Last time we saw how to do automated model selection using the ES function from the smooth package. Now I want to show how to produce combined forecasts from ETS. openforecast.org/2026/04/27/s... #forecasting #datascience #machinelearning #python

As some of you have heard, the smooth package is now on PyPI. So, I’ve decided to write a series of posts showcasing how some of its functions work. We start with the basics, ETS. openforecast.org/2026/04/22/s... #forecasting #datascience #machinelearning

Great news, everyone! After years of development in R, the first release of the smooth forecasting package is now available for Python. Why is this great news? I explain it in the following post: openforecast.org/2026/04/09/s... #forecasting #python #machinelearning #datascience

There’s no such thing as “deterministic forecast”. Sometimes I see people referring to a “deterministic” forecast, and I have some personal issues with this. Because if you apply a model to data then there is nothing deterministic about your forecasts! openforecast.org/2026/03/02/t... #datascience

Apparently, we need to talk about scaling of error measures because this is not as obvious as it seems. openforecast.org/2026/02/23/s... #forecasting #datascience #machinelearning

Great news, everyone! smooth package for R version 4.4.0 is now on CRAN. Why is this a great news? Let me explain! openforecast.org/2026/02/09/s... #forecasting #rstats #datascience #machinelearning

Here is one small, unexpected piece of news: I now have my first package on PyPI! It’s called fcompdata, and let me tell you a little bit about it: openforecast.org/2026/01/26/f... #forecasting #machinelearning #datascience #python

What do you use for model selection? Do you select the best model based on its cross-validated performance, or do you use in-sample measures like AIC? If so, there is a way to improve your selection process further... openforecast.org/2026/01/19/r... #forecasting #machinelearning #datascience

Teaching statistics as a flipped classroom with the help of AI? You heard that right! That’s exactly what I tried this year – and here are the results: openforecast.org/2026/01/07/t... #teaching #statistics

Happy New Year, everyone! 2025 is over (along with the short break we had), and 2026 has arrived. What do I expect from it? Well, I can give you a few hints: ivan.svetunkov.com/en/2026/01/0... #forecasting #personal #plans

When I read posts written by some ML experts, I notice that they either overlook or do not explain a few crucial steps in demand forecasting. In this post, I want to highlight the three most important ones (IMHO). openforecast.org/2025/10/08/s... #forecasting #machinelearning

You might have missed a LinkedIn post about Valeriy Manokhin started by Gael Varoquaux: www.linkedin.com/posts/gael-v... - I've tried to keep dialogue with him for a couple of years, hoping that he would be reasonable. It appears that the whole ML and DS communities know of his toxic behaviour...

Some of you know that the UK education sector is in crisis. The blame is often placed on Chinese students, who have stopped coming to the UK for increasingly expensive education. We have issues at Lancaster University, so I wrote a post: ivan.svetunkov.com/en/2025/07/3... #education #crisis

This year, I couldn’t attend the International Symposium on Forecasting, which I usually do, so instead I went to Gran Canaria for ITISE 2025 to present my research with Nikos Kourentzes on pooling. Read more about it here: openforecast.org/2025/07/21/i... #forecasting #datascience #machinelearning

Good news, everyone! We have finally made some progress translating the smooth package from R to Python (done between Filotas Theodosiou, Leonidas Tsaprounis and me). I've written a short post on that with some links: openforecast.org/2025/06/30/i... #python #datascience #rstats

Are there any PhD students in the crowd who want to learn more about forecasting? What about academic supervisors who have such students? Show me your hands! This post is for you! openforecast.org/2025/06/22/n... #forecasting #predictiveanalytics #course

There's a large area in time series analysis called "changepoint detection". I hadn't worked there before, but thanks to Rebecca Killick and Tom Grundy, I contributed to the paper, which has just been published in the Journal of Time Series Analysis. openforecast.org/2025/06/11/o...

SBC is not for you! I’ve been acting as a reviewer lately, providing comments on papers about intermittent demand, and I’ve felt a bit frustrated by what some authors write. Let me explain. openforecast.org/2025/06/04/s... #machinelearning #datascience #forecasting #intermittent #demand

NHS has a helpful page with a set of questions you can ask your GP to ensure you receive the right treatment for your illness. Surprisingly, these questions can be applied in other fields as well. Here’s an example in applied forecasting, working with companies. openforecast.org/2025/05/28/s...

Fundamental Flaw of the Box-Jenkins Methodology If you've taken a course on forecasting or time series analysis, you’ve probably heard of ARIMA and the Box–Jenkins methodology. IMHO, the methodology has a fundamental flaw and should not be used. Here’s why. openforecast.org/2025/05/13/f...

It was great to attend the 5th IMA & OR conference and present my research with Anna Sroginis on model-based demand classification. openforecast.org/2025/05/02/5... #forecasting #demandplanning #datascience #operationsresearch

The AI2027 exercise (ai-2027.com) is the n-tieth example of Silicon Valley inbued-thinking boys (even when they are not kids) extrapolating trends with back-of-the-envelope calculations to justify their unidimensional sci-fi dreams of AGI. >>

For the last year, Anna Sroginis and I have been working on a paper, modernising demand classification schemes and making them useful in the brave new era of machine learning. I've written a post about this paper, you can read it here: openforecast.org/2025/04/11/s... #machinelearning #forecasting

Challenges related to seasonal data: shifting seasonality There are many different issues with capturing seasonality in time series. In this short post, I’d like to discuss one of the most annoying ones: openforecast.org/2025/04/07/c... #forecasting #datascience #machinelearning #daylight

There is still time to sign up for our forecasting course. See the link below. #forecasting #datascience #machinelearning

If someone tells you that method X solves all problems and is the best one ever, they are either lying or do not fully understand what they are talking about. There is no such thing as “the best approach for everything”. Let me explain: openforecast.org/2025/03/06/t... #forecasting #machinelearning

Who (or what) is "Forecasting academia"? A short post on my website to explain what this term means: openforecast.org/2025/02/27/w...

Over the last few years, I’ve seen phrases like “LightGBM model” or “Neural Network model” on LinkedIn many times, and the statistician in me shivers every time. So, I figured we should discuss that topic... again... #machinelearning #datascience #stats openforecast.org/2025/02/04/m...

Don’t use MAE-based error measures for intermittent demand! A new post for social media on my website: openforecast.org/2025/01/21/d... #forecasting #machinelearning #datascience

There is no such thing as an “assumption-free approach”. One thing that bothers me when I read posts on social media or papers in peer-reviewed journals is the claim that a proposed approach is “assumption-free.” In forecasting, this is never true. Here’s why: openforecast.org/2025/01/07/t...

A paper to read over the Xmas holiday: Wang et al. (2023) – Forecast combinations: An over 50-year review by Xiaoqian Wang, @robjhyndman.com, Feng Li, and Yanfei Kang openforecast.org/2024/12/23/a...

Intermittent demand: don’t try to predict WHEN it will happen: openforecast.org/2024/12/11/i... #forecasting #machinelearning #ml #datascience

A good summary of a very important research by @kandrika.bsky.social

Why Naive is not a good benchmark for intermittent demand? Here is a brief post to answer this question: openforecast.org/2024/12/02/w... #forecasting #machinelearning #datascience

It was a great experience. We are now thinking of repeating this in Spring 2025

Here is an experiment on hacking the TSB model for intermediate time series to accommodate availability constraints. We provide a way to add availability constraints to forecast the "true" demand and avoid just predicting zeros. juanitorduz.github.io/availability...

I've written and recorded a song dedicated to a friend of mine: www.youtube.com/watch?v=MD_D... #music #songs #inmemoryof #johnboylan

Why zeroes happen: a post discussing what is and what isn't intermittent demand and why it is important to tell the difference: openforecast.org/2024/11/18/w... #forecasting #datascience #machinelearning

Reminder that this is happening in one week! 👇👇👇

Are all forecasts wrong? You’ve probably heard the phrase “all forecasts are wrong”, suggesting that the future is unpredictable and that no forecast will ever match the actual outcome. Well, this phrase is not entirely correct, and here’s why: openforecast.org/2024/08/20/a...

Why is it hard to beat the Simple Moving Average in supply chain data? I have summarised some thoughts about that in the following post: openforecast.org/2024/10/28/w... #forecasting

This is funny because it's true.

I thought a forecasters starter pack might be useful, but I can't find many people. This is a start. Please suggest additions. go.bsky.app/NJJseT8

Straight line is just fine. A new post for social media on my website: openforecast.org/2024/09/03/s... #forecasting #MachineLearning #DataScience

Seasonal or not? A short post on my website, discussing what is a natural, and a spurious seasonality: openforecast.org/2024/05/15/s... #forecasting #DataScience #dataanalytics #timeseries

Avoid using MAPE! Here is a post from my website explaining, why: openforecast.org/2024/04/17/a... - the image shows Mean APEs (APE - Absolute Percentage Error). #forecasting #DataScience #demandplanning #MachineLearning

Stop reporting several error measures just for the sake of them! openforecast.org/2024/04/03/s... - a short post on my website. #forecasting #DataScience #MachineLearning #DataAnalytics

Ever heard of ARIMA? It's a key forecasting model in academia, that is not always favored by practitioners. Why? What's wrong with ARIMA? I have written a short post on that topic: openforecast.org/2024/03/21/w... #DataScience #forecasting #MachineLearning