My presentation was based on this paper. And here are the slides:
2025-05-01-IMA-OR

Message 5th IMA and OR Society Conference first appeared on Open Forecasting.

Abstract: Bootstrap is extensively used in statistics and machine learning for cross-sectional data to account for uncertainty about the data, model form, and parameter estimates. However, conventional methods may not be suitable for time series data due to autocorrelation and specific dynamic structures. Over the years, various approaches have been developed to address this issue. Some assume specific models (e.g., STL), while others are non-parametric (e.g., Maximum Entropy Bootstrap, MEB). However, the former can be overly restrictive, while the latter may not perform well in case of outliers and external drivers. To address these issues, we propose a non-parametric bootstrap approach inspired by MEB, which does not assume any structure in the data yet creates reasonable copies of existing time series of different nature. These copies can be utilised in bagged ETS/ARIMA or any other approach involving small sample uncertainty. We demonstrate how the proposed bootstrap works using real-time series examples and assess improvements it brings in terms of forecasting accuracy compared to conventional approaches.

Here are the slides of the presentation.

And here is me, trying not to attract attention of statisticians:

How to attract attention of statisticians…

Message ISF2024: How to Bootstrap Time Series without Attracting Attention of Statisticians first appeared on Open Forecasting.

ETS+ARIMA love story with happy ending…

This talk was based on Chapter 9 of ADAM monograph, and more specifically on Section 9.4.

The slides of the presentation are available here.

Message ISF2022: How to make ETS work with ARIMA first appeared on Open Forecasting.

Presentation at ISF2021 Lancaster Hub

I presented on the topic of Multiplicative ETS, based on this chapter of the ADAM textbook and on the paper John Boylan and I are working on. In this presentation, I discuss that the point forecasts from the ETS(M,*,*) models in general do not correspond to conditional expectations (and not even to geometric means) and show the difference between the two. Furthermore, we propose using Log Normal, Gamma and Inverse Gaussian distributions for the error term of the model instead of the Normal one. This makes the model more realistic and help in forecasting on low volume data. All of this is already implemented in adam() function of smooth package for R.

Here are the slides.

Message ISF2021: How to Make Multiplicative ETS Work for You first appeared on Open Forecasting.

Retail sales of tomatoes

Anyway, here are the slides if anyone wants to have a look.

Message International Symposium on Forecasting 2019 first appeared on Open Forecasting.

1. State space ARIMA for Supply Chain Forecasting, based on which I have developed a module for Smoothie a couple of years ago,
2. Artificial Intelligence in Business, one of the modern hot topics that the company wanted to know a little bit more about.

Presentation at SMUG2019

The conference was interesting, showing what the company does and what it stands for. They are doing a good job in developing the software for forecasting and inventory control and supporting their users. Plus, I finally had a chance to meet in person with both founders of the company (Bill Tonetti and Eric Townson), as well as with the other members of their team. Overall, it was a pleasant experience and an interesting event.

As for the presentations, they seemed to go well, and the participants of the conference looked satisfied. Here are the slides:

1. SMUG2019 – Svetunkov – ARIMA
2. SMUG2019 – Svetunkov – AI in Business

Message SMUG2019 first appeared on Open Forecasting.

1. 5th April, CMAF Workshop on Retail Demand Forecasting and Data Science, London UK. This is an event organised by the Centre for Marketing Analytics and Forecasting of Lancaster University, UK. This is focused on practitioners. Anyone interested can register on the website and attend, this is a free event.
2. 29th April, CMAF Workshop on Business Forecasting using R: Are we there yet? London, UK. This is also organised by CMAF and this is also a free event. The workshop might be interesting to practitioners who want to know more about R, how to use it for decision making and whether it is worth investing in support systems based on R.
3. 23 – 24th May, 25th IIF Workshop: Predictive Analytics: Theory, Applications and Algorithms, Cambridge, UK. This is the workshop organised by the International Institute of Forecasters. It should be interesting to the forecasting and machine learning audiences.
4. 3 – 5th September, Forecasting stream at OR61, Kent University, UK. The stream is organised by Robert Fildes and John Boylan. It’s too early to say who specifically will be there, but the OR conferences are usually quite versatile and are interesting to attend.

You can still attend the following events:

1. 16 – 19th June, 39th International Symposium on Forecasting in Thessaloniki, Greece. I am presenting on the topic of intermittent demand and also co-authoring three other talks (on vector exponential smoothing, on hierarchical forecasting and on complex-valued modelling).
2. 20 – 22nd June, ISMS Marketing Science Conference 2019, Rome, Italy. I won’t attend this myself, but some of my colleagues are going. This should be an interesting event, taking place in a great city.
3. 9 – 12th July, useR!, Toulouse, France. I will be presenting the functions of smooth package and I will be giving the hex stickers for smooth and greybox packages for free at this event. So if you have spaces on your laptop that you don’t know how to fill in, going to Toulouse and getting hexes is the best thing you can do this summer…

Hopefully, I will be able to come up with the next post like this some time before the end of 2019.

Message Conferences 2019 first appeared on Open Forecasting.

Here’s the abstract:

Exponential smoothing has been known in both theoretical and practical forecasting for more than 60 years. It has evolved substantially from a simple exponential smoothing method, aiming at dealing with level data to a state-space framework, covering various time series characteristics. In this presentation we discuss the key milestones in the development of exponential smoothing, show the connections between the exponential smoothing and the other forecasting models and, finally, propose a more general framework that can potentially encompass all the existing forecasting models, called “Generalised Univariate Model”.

And here are the slides

Message OR60 presentation. Forecasting using exponential smoothing: the past, the present, the future first appeared on Open Forecasting.

Here are the slides from the presentation. My colleagues and I are working on a paper for this. But this is related to the paper John and I submitted to IJF.

An example of time series discussed in the presentation

There is also now a video of my presentation:

Message International Symposium on Forecasting 2018 first appeared on Open Forecasting.

• International Symposium on Forecasting, 17-20 June, 2018, Boulder, Colorado, USA. This is the key event in forecasting, which any self-respecting forecaster should not miss. I don’t know what to present there yet, but I’ll come up with something as an excuse to attend the symposium.
• Informs Society for Marketing Science Conference, 13-16 June, 2018, Temple University, Fox School of Business, Philadelphia, US. This conference will be held week before ISF2018, so why not kill to birds with one stone (or at least scare them). I’m going to present a topic, I’ve been working on with my colleagues Victoria Grigorieva (Higher School of Economics, Saint Petersburg, Russia) and Yana Salihova (Economic State University, Saint Petersburg, Russia).
• OR60, 11–13 September, 2018, Lancaster, UK. OR conference this year will be held in Lancaster, and there will be a forecasting stream (or maybe several), so it would be shame to miss it.
• useR!, 10-13 July, 2018, Brisbane, Australia. This is a specialised R conference, where I might present my smooth package for R. However Australia is far and expensive, so I’m not yet sure if I would go there. But it’s good to keep this event in mind.

Finally, there are so called “Quarterly Forecasting Forums” happening 4 times a year in England. They are organised by an unofficial consortium of universities (currently only four): Bath, Cardiff, Coventry and Lancaster. The next event is planned for the end of February – beginning of March in Lancaster. After that the forum will be held in Bath. Unfortunately, I don’t yet have any details about these events, so stay tuned. If you are interested in participating, please, send me a message or leave a comment to this post.

UPDATE: Bahman Rostami-Tobar pointed out to me that there will be a 24th IIF Workshop on “Forecasting for Social Good“, 12-13 July, 2018 in Cardiff, UK. Have a look, this should be an interesting event.

Message Conferences 2018 first appeared on Open Forecasting.