Similarly to this question I am looking for time series data sets that show some extent of change in the feature distribution(s) over time.
The requirements that I have regarding the data set are that it has:
- measurements of some numerical feature(s) over time
- more than one measurement per time point
- some change in the distribution over time, so the distribution in the beginning is different from that at the end (timewise)
- (ideally) a multimodal feature distribution (=multiple peaks)
- More than 10k samples
Besides this there are no restrictions regarding aspects like the region, origin or context of the data.
There are many stock exchange and energy grid load data sets that simply have one measurement per time point, so these aren't really relevant to me.
I was looking around a bit and I figured that maybe environmental or biological data sets might fit, for example the concentration of a compound in sampled plants or something like that.
I'd be thankful for suggestions, references or even just keywords that might help me find such data.