Forecasting in Python with Prophet

Re:Forecasting in Python with Prophet by maliker on Friday February 22, 2019 @02:09PM (#19228141)

We've enjoyed using Prophet for forecasting in our work with electric utilities. It's pretty fire-and-forget. We wish it had an option for fast retraining of the model when we get new hourly data, though. That's pushing us towards other options in scipy.

[ Parent | on HN ]

Re:Forecasting in Python with Prophet by wenc on Friday February 22, 2019 @02:33PM (#19228370)

I looked at Prophet a few months ago because we needed a fire-and-forget library similar to 'auto.arima' (R "forecast" package) for Python, but no good candidates existed.

However I found Prophet to be computationally a little heavier than auto.arima because it uses "stan" (Bayesian) underneath, which in turn uses an MCMC type approach and has quite a few dependencies. We needed fast model retraining as well, and at the time, it didn't seem like that was something it excelled at. (might have changed, I'm not sure)

I ended up putting together a simple ensemble forecast model class with "statsmodels" which automatically selected/averaged the best models over a collection of model types via heuristics and cross-validation. It works ok, but I'm still waiting for someone to port the R auto.arima over to Python. (I tried rpy, which in theory should have worked, but I struggled with the impedance mismatch)

[ Parent | on HN ]

Re:Forecasting in Python with Prophet by venuur on Friday February 22, 2019 @11:19PM (#19231591)

I found this port of auto-arima for Python. I haven’t used it in production, but it was easy to test on some demo data. https://pypi.org/project/pyramid-arima/

[ Parent | on HN ]

Re:Forecasting in Python with Prophet by wenc on Sunday February 24, 2019 @08:44PM (#19242072)

Thanks, I'll take a look.

[ Parent | on HN ]

Re:Forecasting in Python with Prophet by Tarq0n on Saturday February 23, 2019 @07:18PM (#19236273)

You only need to do MCMC if you want simulated confidence intervals, it doesn't do anything for model estimation.

[ Parent | on HN ]

Re:Forecasting in Python with Prophet by wenc on Sunday February 24, 2019 @09:34AM (#19238679)

You’re right. It uses Stan to do MAP for model estimation and HMC for (optional) computation of confidence intervals. [1]

Thanks for the correction.

I need to look at Prophet again. Maybe it will work for fast model retraining. I had trouble the last time I tried it.

[1] https://research.fb.com/prophet-forecasting-at-scale/

[ Parent | on HN ]