Highly periodical time series datasets?

Question

I'm looking for a highly periodical time series dataset. Mainly, this is what I am looking for:

Periodic time series. No preference over the frequency, it can be annual, monthly or even daily.
The time series can be fine even in case it is periodical in some parts of it but not periodical in general
At least 100 observations would be nice
The dataset should be free to use to everyone and publicly available.

Could you please provide more details, to help us answer: I am not sure of what you mean by "highly periodical" wrt "even in case it is periodical in some parts of it but not periodical in general"? What is the purpose, in a few words? — Laurent Duval, Aug 27 '16 at 16:31
Does this answer your question? Seasonal Sales Data for time Series Analysis (I don't have enough rep here to VTC as a duplicate.) — Stephan Kolassa, Jun 14 '21 at 09:04

score 2 · Answer 1 · answered May 30 '17 at 18:56

If you need something purely periodic, consider using sunrise/sunset for a given latitude/longitude.

There are simple scripts available to generate as much data as you need. See this Stackoverflow thread for some python and java options. Python implementations oftenuse the Ephem package.

They also mention using a web-service, sunrise-sunset.org.

score 1 · Answer 2 · answered May 30 '17 at 17:09

In Germany water levels are published by the German "Wasserstraßen und Schifffahrtsverwaltung" which is part of the Federal Ministry of Transport and Digital Infrastructure.

They offer REST-APIs free of charge and for inclusion in apps.

Check for example water levels for Helgoland (in the North Sea)

JSON: https://www.pegelonline.wsv.de/webservices/rest-api/v2/stations/HELGOLAND%20SÜDHAFEN/W/measurements.json?start=P15D

API Documentation is here (in German).

There will be offerings from similar organisations in other countries.