News for hackers. Stuff that matters.
Re:Facebook Prophet: library for generating forecasts from a... by ImaCake on Tuesday September 26, 2023 @07:07PM (#37667086)
> they expect that your data comes at a very regular interval

Does prophet rely on this assumption? For health timeseries data the tool of choice is survival analysis - typically using Cox proportional hazards regression or similar regression tools that are able to handle irregular or censored data.

I've seen some moves towards using fancy bayesian or fancier machine learning stuff for clinical trials but a big issue is that they are very difficult to communicate to their intended audience.

Re:Facebook Prophet: library for generating forecasts from a... by cjalmeida on Tuesday September 26, 2023 @07:41PM (#37667471)
I find Bayesian regression models are actually simpler to explain as the assumptions you make are explicit and part of the model specification.

(Thought the actual sampling mechanics and tooling can be much more complex)

Re:Facebook Prophet: library for generating forecasts from a... by carlossouza on Tuesday September 26, 2023 @11:28PM (#37669637)
PyMC or Pyro/NumPyro make the implementation of Bayesian regression dead simple
Re:Facebook Prophet: library for generating forecasts from a... by rdli on Tuesday September 26, 2023 @07:53PM (#37667588)
I tried Prophet via Darts, and all the models in Darts assume a regular time series.

Re: "fancier machine learning" -- I've seen different flavors of RNNs & LSTMs have some success in analyzing time series data. I've struggled to get them to work on real-world (i.e., messy) data, but have had some encouraging results with a transformer encoder-only NN.

Re:Facebook Prophet: library for generating forecasts from a... by laichzeit0 on Wednesday September 27, 2023 @12:24AM (#37670101)
There’s also the auton survival library. I’ve used it for very big survival models with time varying coefficients:

https://autonlab.org/auton-survival/