Comments for Open Forecasting

Comment on Intermittent demand classifications: is that what you need? by Ivan Svetunkov

Ivan Svetunkov — Sun, 14 Sep 2025 13:49:46 +0000

Sorry for not replying earlier, somehow I missed this comment.

For me, the question is what you are going to do with that afterwards? If you want to use that to choose between different forecasting techniques, maybe just a simple split to “have zeroes/doesn’t have” would suffice. If it is just a curiosity, then who cares how you categorise the demand?

Also, have a look at this: https://openforecast.org/2025/04/11/svetunkov-sroginis-2025-model-based-demand-classification/ – I’ve been working with Anna Sroginis on an alternative classification. The main idea is to first understand whether the zeroes happen naturally or artificially and then do a classification to regular/intermittent (with lumpy/smooth inside). This paper is under review right now, when it gets published, I’ll write more about it (it has already changed a bit).

Comment on Intermittent demand classifications: is that what you need? by Kishor Kukreja

Kishor Kukreja — Mon, 28 Jul 2025 22:33:59 +0000

More often I have seen the usage of ADI and CV to classify the demand type into 4 buckets.
How would you suggest one can go about understanding what the demand type of a product is ? If the goal is to only understand the type of demand for products at scale, is there a different way ?

Comment on Who is “Forecasting academia”? by Ivan Svetunkov

Ivan Svetunkov — Thu, 27 Mar 2025 09:31:52 +0000

In reply to Mohamed Merabtine.

Hi Mohamed,

The list is pretty subjective, I don’t know the work of Eric Ghysels, but thank you for the suggestion!

And thanks for the kind words!

Comment on Who is “Forecasting academia”? by Mohamed Merabtine

Mohamed Merabtine — Wed, 26 Mar 2025 19:00:39 +0000

My list would include Eric Ghysels for his contributions to the modelling and forecasting of mixed frequency data.

PS: I came from LinkedIn. Thanks for pointing out to the existence of your blog, I enjoy reading your posts!

Comment on Why Naive is not a good benchmark for intermittent demand by Ivan Svetunkov

Ivan Svetunkov — Thu, 09 Jan 2025 09:44:21 +0000

Yes, sure. I would include the following in the list of benchmarks for intermittent demand:

1. SMA,
2. Simple Exponential Smoothing,
3. Global Average,
4. Zero forecast (just a sanity check to make sure that everything is in order).

The more complicated benchmarks would be:
5. Croston’s method (https://doi.org/10.2307/3007885)
6. TSB (https://doi.org/10.1016/j.ejor.2011.05.018)

Hope that helps.

Comment on Why Naive is not a good benchmark for intermittent demand by Indar Kusmadi

Indar Kusmadi — Thu, 09 Jan 2025 03:20:19 +0000

The blog post mentions SMA as a possible alternative to Naive for forecasting intermittent demand. Could you elaborate on other forecasting methods that might be more suitable for this scenario? Telkom University Jakarta

Comment on John E. Boylan by Indar Kusmadi

Indar Kusmadi — Tue, 07 Jan 2025 02:26:52 +0000

The article beautifully describes your personal and professional relationship with John Boylan. Could you share a specific anecdote that best exemplifies his mentorship and the impact he had on your academic and professional growth? Telkom University Jakarta

Comment on smooth & greybox under LGPLv2.1 by Ivan Svetunkov

Ivan Svetunkov — Thu, 09 May 2024 15:38:07 +0000

In reply to Eric Groo. Hi Eric, Thanks for getting in touch with me! I've sent you an email regarding this.

Comment on smooth & greybox under LGPLv2.1 by Eric Groo

Eric Groo — Wed, 08 May 2024 20:55:48 +0000

Hi – I would love to help with the python development. I’ve been pitching your methods in my office for a while, and folks are interested in a python implementation. How can you put me in contact with your colleagues to help?

Comment on smooth v3.2.0: what’s new? by Ivan Svetunkov

Ivan Svetunkov — Sat, 11 Feb 2023 22:21:28 +0000

In reply to Mark Neal.

Hi Mark,

I only judge by LinkedIn posts and general rise of popularity of Python over R in business. Python’s Statistical methods seem to be lagging behind the R ones, but I think it will change soon. Nonetheless, I plan to continue supporting my packages anyway :).

Cheers,
Ivan