As for my contribution, I presented a paper that Nikos Kourentzes and I have been working on since around 2018. It focuses on pooling using point information criteria. The core idea is to combine forecasts based on a smaller pool of models, which we propose creating by comparing the distributions of information criteria across forecasting models. We’re planning to finish a new version of the paper by September and submit it to a peer-reviewed journal. I’ll share more details when the draft that I can share is ready. In the meantime, you can check out the slides that summarise the main points of the paper. Here they are.

Message ITISE2025: Beyond summary performance metrics for forecast selection and combination first appeared on Open Forecasting.

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.

Message Why do zeroes happen? A model-based view on demand classification 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.

• There has been a huge progress in the area of Exponential Smoothing for the last 40 years. This includes development of state space Single Source of Error model by Ralph Snyder, Keith Ord, Anne Koehler and Rob Hyndman, which is well summarised in the book of Hyndman et al. (2008). This also includes development of TBATS by de Livera et al. (2011), MAPA by Kourentzes et al. (2014) and many other things, including some parts from my monograph on ADAM;
• No, Exponential Smoothing is not a special case of ARIMA. This is discussed, for example, here and here;
• Yes, Exponential Smoothing can handle external information, ETSX works fine and can be used efficiently in practice. This was shown, for example, by Kourentzes & Petropoulos (2016), Ramos et al. (2023) and even in M5 competition (ETSX did better than the plain ETS by roughly 6%);
• The modern Exponential Smoothing can handle intermittent demand and/or multiple frequencies. It can be estimated using multistep losses and regularisation (see Pritularga et al, 2023);
• If you decide to use Exponential Smoothing you should use the modern form of it. Do not ignore all the hard work of Hyndman et al. (2008) and related research. So, you should not use this formulation:
  
  

  Outdated formulation of exponential smoothing

  It is outdated and shows that you have completely ignored all the developments in the area since 1985 (side note: when reviewing papers, if I see that authors use this formulation, I automatically flag the paper as a major revision). You should use the modern approach instead, State Space Single Source of Error model, i.e. this formulation:
  \begin{equation*}
  \begin{aligned}
  & {y}_{t} = l_{t-1} + b_{t-1} + s_{t-m} + \epsilon_t \\
  & l_t = l_{t-1} + b_{t-1} + \alpha \epsilon_t \\
  & b_t = b_{t-1} + \beta \epsilon_t \\
  & s_t = s_{t-m} + \gamma \epsilon_t
  \end{aligned} .
  \end{equation*}

  If you see someone using the old formulation, know that they do not know the state-of-the-art forecasting.

Here are the slides of the presentation.

And here is the recording of the webinar:

Message Why you should care about Exponential Smoothing 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.

The slides of the presentation are available here.

Message Multi-step Estimators and Shrinkage Effect in Time Series Models – presentation for CEBA first appeared on Open Forecasting.

Today I’ve presented my smooth package at the useR!2019 conference in Toulouse, France. This is a nice conference, focused on specific solutions to specific problems. Here, people tend to present functions from their packages (not underlying models, like, for example, at ISF). On one hand, this has its own limitations, but on the other hand, this is nice, because attending the conference, one can find out, what specific solutions there are for different problems. For example, now I know what packages I can use for the anomaly detection or text mining. This is quite handy for the practising analysts.

The presentation itself seemed to have gone okay, although I did not feel comfortable standing in one place and talking in the microphone. They restricted my freedom! But this was necessary in order to record the presentation… I would prefer to roam free and shout rather than stand in one place and use microphone… But that’s the price that one has to pay :).

The slides from the presentation can be downloaded from here.

UPDATE: The video of the presentation is now available on the R Consortium YouTube channel:

Message useR!2019 in Toulouse, France 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.