Retime

The retime function allows you to retime, i.e. change the timestamps of a TimeArray, similar to what Matlab's retime does.

using Plots, Dates, TimeSeries
gr()
timestamps = range(DateTime(2020, 1, 1), length = 7*24, step = Hour(1))
ta = TimeArray(timestamps, cumsum(randn(7*24)), [:a])

168×1 TimeArray{Float64, 1, DateTime, Vector{Float64}} 2020-01-01T00:00:00 to 2020-01-07T23:00:00
┌─────────────────────┬──────────┐
│                     │ a        │
├─────────────────────┼──────────┤
│ 2020-01-01T00:00:00 │ 0.945461 │
│ 2020-01-01T01:00:00 │ -0.64693 │
│ 2020-01-01T02:00:00 │ -3.65944 │
│ 2020-01-01T03:00:00 │ -4.57012 │
│ 2020-01-01T04:00:00 │ -5.56053 │
│ 2020-01-01T05:00:00 │ -6.63431 │
│ 2020-01-01T06:00:00 │  -6.9126 │
│ 2020-01-01T07:00:00 │ -8.66276 │
│          ⋮          │    ⋮     │
│ 2020-01-07T17:00:00 │ -7.25812 │
│ 2020-01-07T18:00:00 │ -6.84261 │
│ 2020-01-07T19:00:00 │ -6.18256 │
│ 2020-01-07T20:00:00 │ -6.93157 │
│ 2020-01-07T21:00:00 │ -6.22709 │
│ 2020-01-07T22:00:00 │ -5.82314 │
│ 2020-01-07T23:00:00 │ -7.31261 │
└─────────────────────┴──────────┘
                  153 rows omitted

Using a new time step

retime(ta, Minute(15))

669×1 TimeArray{Float64, 2, DateTime, Matrix{Float64}} 2020-01-01T00:00:00 to 2020-01-07T23:00:00
┌─────────────────────┬──────────┐
│                     │ a        │
├─────────────────────┼──────────┤
│ 2020-01-01T00:00:00 │ 0.945461 │
│ 2020-01-01T00:15:00 │ 0.945461 │
│ 2020-01-01T00:30:00 │ 0.945461 │
│ 2020-01-01T00:45:00 │ 0.945461 │
│ 2020-01-01T01:00:00 │ -0.64693 │
│ 2020-01-01T01:15:00 │ -0.64693 │
│ 2020-01-01T01:30:00 │ -0.64693 │
│ 2020-01-01T01:45:00 │ -0.64693 │
│          ⋮          │    ⋮     │
│ 2020-01-07T21:30:00 │ -6.22709 │
│ 2020-01-07T21:45:00 │ -6.22709 │
│ 2020-01-07T22:00:00 │ -5.82314 │
│ 2020-01-07T22:15:00 │ -5.82314 │
│ 2020-01-07T22:30:00 │ -5.82314 │
│ 2020-01-07T22:45:00 │ -5.82314 │
│ 2020-01-07T23:00:00 │ -7.31261 │
└─────────────────────┴──────────┘
                  654 rows omitted

Using new timestep vector

new_timestamps = range(DateTime(2020, 1, 1), DateTime(2020, 1, 2), step = Minute(15))
retime(ta, new_timestamps)

97×1 TimeArray{Float64, 2, DateTime, Matrix{Float64}} 2020-01-01T00:00:00 to 2020-01-02T00:00:00
┌─────────────────────┬──────────┐
│                     │ a        │
├─────────────────────┼──────────┤
│ 2020-01-01T00:00:00 │ 0.945461 │
│ 2020-01-01T00:15:00 │ 0.945461 │
│ 2020-01-01T00:30:00 │ 0.945461 │
│ 2020-01-01T00:45:00 │ 0.945461 │
│ 2020-01-01T01:00:00 │ -0.64693 │
│ 2020-01-01T01:15:00 │ -0.64693 │
│ 2020-01-01T01:30:00 │ -0.64693 │
│ 2020-01-01T01:45:00 │ -0.64693 │
│          ⋮          │    ⋮     │
│ 2020-01-01T22:30:00 │ -2.54295 │
│ 2020-01-01T22:45:00 │ -2.54295 │
│ 2020-01-01T23:00:00 │ -2.72574 │
│ 2020-01-01T23:15:00 │ -2.72574 │
│ 2020-01-01T23:30:00 │ -2.72574 │
│ 2020-01-01T23:45:00 │ -2.72574 │
│ 2020-01-02T00:00:00 │ -1.71528 │
└─────────────────────┴──────────┘
                   82 rows omitted

You can perform retime on irregularly spaced timestamps, both using a TimeArray with irregular timestamps or using a vector of irregular timestamps. Depending on the timestamps upsampling or downsampling is used.

new_timestamps = vcat(
    range(DateTime(2020, 1, 1), DateTime(2020, 1, 2)-Minute(15), step = Minute(15)),
    range(DateTime(2020, 1, 2), DateTime(2020, 1, 3), step = Hour(1)),
)
retime(ta, new_timestamps)

121×1 TimeArray{Float64, 2, DateTime, Matrix{Float64}} 2020-01-01T00:00:00 to 2020-01-03T00:00:00
┌─────────────────────┬────────────┐
│                     │ a          │
├─────────────────────┼────────────┤
│ 2020-01-01T00:00:00 │   0.945461 │
│ 2020-01-01T00:15:00 │   0.945461 │
│ 2020-01-01T00:30:00 │   0.945461 │
│ 2020-01-01T00:45:00 │   0.945461 │
│ 2020-01-01T01:00:00 │   -0.64693 │
│ 2020-01-01T01:15:00 │   -0.64693 │
│ 2020-01-01T01:30:00 │   -0.64693 │
│ 2020-01-01T01:45:00 │   -0.64693 │
│          ⋮          │     ⋮      │
│ 2020-01-02T18:00:00 │    1.26957 │
│ 2020-01-02T19:00:00 │  -0.660237 │
│ 2020-01-02T20:00:00 │   -1.51387 │
│ 2020-01-02T21:00:00 │   -1.68267 │
│ 2020-01-02T22:00:00 │   0.458611 │
│ 2020-01-02T23:00:00 │ -0.0921172 │
│ 2020-01-03T00:00:00 │  -0.286219 │
└─────────────────────┴────────────┘
                    106 rows omitted

Upsampling

Interpolation is done using the upsample argument. If no data is directly hit, the specified upsample method is used. Available upsample methods are:

Linear() or :linear
Nearest() or :nearest
Previous() or :previous
Next() or :next

ta_ = retime(ta, Minute(15), upsample=Linear())

669×1 TimeArray{Float64, 2, DateTime, Matrix{Float64}} 2020-01-01T00:00:00 to 2020-01-07T23:00:00
┌─────────────────────┬───────────┐
│                     │ a         │
├─────────────────────┼───────────┤
│ 2020-01-01T00:00:00 │  0.945461 │
│ 2020-01-01T00:15:00 │  0.547363 │
│ 2020-01-01T00:30:00 │  0.149265 │
│ 2020-01-01T00:45:00 │ -0.248832 │
│ 2020-01-01T01:00:00 │  -0.64693 │
│ 2020-01-01T01:15:00 │  -1.40006 │
│ 2020-01-01T01:30:00 │  -2.15318 │
│ 2020-01-01T01:45:00 │  -2.90631 │
│          ⋮          │     ⋮     │
│ 2020-01-07T21:30:00 │  -6.02512 │
│ 2020-01-07T21:45:00 │  -5.92413 │
│ 2020-01-07T22:00:00 │  -5.82314 │
│ 2020-01-07T22:15:00 │  -6.19551 │
│ 2020-01-07T22:30:00 │  -6.56787 │
│ 2020-01-07T22:45:00 │  -6.94024 │
│ 2020-01-07T23:00:00 │  -7.31261 │
└─────────────────────┴───────────┘
                   654 rows omitted

plot(ta)
plot!(ta_)

Downsampling

Downsampling or aggregation is done using the downsample argument. This applies a function to each interval not including the right-edge of the interval. If no data is present in the interval the specified upsample method is used. Available downsample methods are:

Mean() or :mean
Min() or :min
Max() or :max
Count() or :count
Sum() or :sum
Median() or :median
First() or :first
Last() or :last

ta_ = retime(ta, Hour(6), downsample=Mean())

28×1 TimeArray{Float64, 2, DateTime, Matrix{Float64}} 2020-01-01T00:00:00 to 2020-01-07T18:00:00
┌─────────────────────┬────────────┐
│                     │ a          │
├─────────────────────┼────────────┤
│ 2020-01-01T00:00:00 │   -3.35431 │
│ 2020-01-01T06:00:00 │   -9.24263 │
│ 2020-01-01T12:00:00 │   -7.98966 │
│ 2020-01-01T18:00:00 │   -3.81945 │
│ 2020-01-02T00:00:00 │  -0.940683 │
│ 2020-01-02T06:00:00 │  -0.246761 │
│ 2020-01-02T12:00:00 │    1.76441 │
│ 2020-01-02T18:00:00 │  -0.370119 │
│          ⋮          │     ⋮      │
│ 2020-01-06T06:00:00 │   -4.10185 │
│ 2020-01-06T12:00:00 │   -4.64036 │
│ 2020-01-06T18:00:00 │   -3.69262 │
│ 2020-01-07T00:00:00 │ -0.0570419 │
│ 2020-01-07T06:00:00 │   -1.35879 │
│ 2020-01-07T12:00:00 │    -6.0298 │
│ 2020-01-07T18:00:00 │   -6.55326 │
└─────────────────────┴────────────┘
                     13 rows omitted

plot(ta)
plot!(ta_)

Extrapolation

Extrapolation at the beginning and end of the time series is done using the extrapolate argument. Available extrapolate methods are:

FillConstant(value) or :fillconstant
NearestExtrapolate() or :nearest
MissingExtrapolate() or :missing
NaNExtrapolate() or :nan

new_timestamps = range(DateTime(2019, 12, 31), DateTime(2020, 1, 2), step = Minute(15))
ta_ = retime(ta, new_timestamps, extrapolate=MissingExtrapolate())