TimeSeries.jl and upsampling

Specific Domains Data
1

Hi,

I need to be able to resample time series both ways : downsampling and upsimpling.
I was able to downsample my time series using both TimeSeries.jl and TimeSeriesResampler.jl
But the later doesn’t support upsampling (mostly using nearest or left or right or linear interpolation).

Has anybody found a solution to that ? Maybe there is a better package for that ?
Any idea ?

Thank you for your help.

2

Maybe Interpolations.jl is useful for your use case, but I haven’t checked whether it composes with Timeseries.jl.

3

DSP.resample will certainly downsample, not sure about upsample.

4

I stumbled upon this problem too, but haven’t found any solution so far:

  • Here people suggest Interpolations.jl, but it seems to only work when time is a real value, not a date.
  • Here TimeSeriesResampler.jl is recommended, but it seems unmaintained and was last updated 5 years ago.
  • There’s Resampling.jl, but it’s unmaintained and was last updated 9 years ago.
  • DataFrames.jl doesn’t seem to support indexing by date/time and it seems like the devs don’t plan to implement this.

Python’s Pandas has DataFrame.resample. Do no Julia packages provide such functionality?