Time series extrema refers to the minimum and maximum values in time series data. It can be useful to compute aggregates of a time series around extrema values to understand trends around those values.
Extrema extraction recipe¶
The extrema extraction recipe allows you to extract aggregates of time series values around a global extremum (global maximum or global minimum).
Using this recipe, you can find a global extremum in one dimension of a time series and perform windowing functions around the timestamp of the extremum on all dimensions. See Windowing for more details about the windowing operations.
This recipe automatically works on all numerical columns (int or float) in your time series data.
Data that consists of equispaced n-dimensional time series in wide or long format.
If input data is in the long format, then the recipe will find the extremum of each time series in the column on which you operate. See Algorithms for more details.
Name of the column that contains the timestamps. Note that the timestamp column must have the date type as its meaning (detected by DSS), and duplicate timestamps cannot exist for a given time series.
Find extremum in column¶
Name of column from which to extract the extremum value.
Type of extremum to find, specified as “Global minimum” or “Global maximum”.
Option to use a causal window, that is, a window which contains only past (and optionally, present) observations. The timestamp for the extremum point will be at the right border of the window.
If you de-select this option, Dataiku DSS uses a bilateral window, that is, a window which places the timestamp for the extremum point at its center.
Window shape applied to the Sum and Average operations. The shape is specified as one of these values:
- Rectangular: simple rectangular window with a flat profile
- Triangle: triangle window (with nonzero values at the endpoints)
- Bartlett: triangle window (with zero values at the endpoints)
- Gaussian: nonlinear window in the shape of a Gaussian distribution
- Parzen: nonlinear window made of connected polynomials of the third degree
- Hamming: nonlinear window generated as a sum of cosines (trigonometric polynomial of order 1)
- Blackman: nonlinear window generated as a sum of cosines (trigonometric polynomial of order 2)
Width of the window, specified as a numerical value (int or float).
The window width cannot be smaller than the frequency of the time series. For example, if your timestamp intervals equal 5 minutes, then you cannot specify a window width that is smaller than 5 minutes.
Unit of the window width, specified as one of these values:
Include window bounds¶
Edges of the window to include when computing aggregations. This parameter is active only when you use a causal window. Choose from one of these values:
- Yes, left only
- Yes, right only
- Yes, both
Operations to perform on a window of time series data. Select one or more of these options:
- Standard deviation
- 25th percentile
- 75th percentile
- First order derivative
- Second order derivative
Column with identifier¶
Name of column that contains identifiers for the different time series in the input data. This applies when your input data is in the long format.
Data consisting of the results of extrema extraction, one row for each time series. Each row contains the timestamp of the extremum and the computed aggregations for a window of data around the extremum.
If the input data is in the wide format, the recipe works as follows:
If the input data is in the long format, then the recipe implements slightly different steps, as follows:
- find the global extremum and corresponding timestamp for each time series in a specific column.
- for all columns, apply a window around the timestamps found in step 1 and then compute aggregations.
- If you have irregular timestamp intervals, first resample your data, using the resampling recipe. Then you can apply the extrema extraction recipe to the resampled data.
- The extrema extraction recipe automatically works on all numerical columns. If you want to use the recipe on select columns, first create a recipe to extract the columns that you plan to use.