Question: unequally spaced timeseries

First, big thanks for a nice library! Very useful!

I am trying to follow your Quick Start for SeasonalAD, but I am encountering a problem. My timeseries seem be to unequally spaced (e.g. 09:05, 09:15, 9:30, 9:55). Hence the SeasonalAD complains: RuntimeError: Series does not follow any known frequency (e.g. second, minute, hour, day, week, month, year, etc. How to overcome this? I have tried rounding my series to 15min, removing duplicates and resampling.

s_train.index = s_train.index.round('15min')
s_train = s_train[~s_train.index.duplicated()]
s_train = s_train.asfreq('15min')

Obviously nothing worked. Any ideas how to solve this? I wish to retain as much granularity as possible.

[zhj0513]

try s_train.resample(‘15min’).ffill() if adtk.data doesn’t have resample

[tailaiw]

@ajdapretnar Seasonal decomposition in ADTK requires the input follows equally spaced time series, as you noticed. Therefore, the time series should be resampled with a constant frequency, for example 15 min.

If you got a ValueError regarding NaN values, the reason is that your resampling may introduce NaN value to the new time series (in your example, 9:45 will have NaN value). Currently, SeasonAD does not support time series with NaN, unless the NaN values are on the starting or ending part of the time series and they will be ignored.

In the adtk.data module, we offer a resample function, which resamples a time series with user-given space. The resampling is based on linear interpolation. You may try it like the follows.

from adtk.data import resample
s_train = resample(s_train, dT="15 min")

If you want to fill NaN with forward or backward filling instead of interpolation, I believe you can also use the fillna method of Pandas.