Time series trends
Examples disclosed herein relate, among other things, to a method. The method may obtain a time series comprising a plurality of data points associated with a sub-segment of a segment, obtaining a plurality of weights associated with a plurality of data point pairs from the plurality of data points, and based on the plurality of weights and the plurality of data point pairs, determine whether the time series comprises a trend. Based on a determination that the time series comprises a trend, the method may calculate a trend score for the trend based on at least one characteristic of at least one of the segment and the sub-segment, and provide the trend for display.
1. A computing device comprising:
a processor; and
a memory to store instructions that, when executed by the processor, cause the processor to:
obtain a time series comprising a plurality of data points associated with a sub-segment of a segment;
obtain a plurality of weights associated with a plurality of data point pairs from the plurality of data points;
determine a stability significance of the time series based on the plurality of weights and the plurality of data point pairs, wherein determining the stability significance of the time series comprises determining a weighted sum of a plurality of signs corresponding to differences of the plurality of data point pairs;
based on the stability significance of the time series, determine whether the time series comprises a trend;
calculate a trend score for the trend based on the stability significance of the time series and at least one characteristic of at least one of the segment and the sub-segment; and
provide data to display the time series on a graphical user interface (GUI).
2. The computing device of claim 1 , wherein the instructions, when executed by the processor, further cause the processor to calculate the trend score based on at least one of:
a discrepancy between the stability significance of the time series and stability significances of a plurality of other time series associated with the segment;
a discrepancy between a stability significance of the segment and stability significances of a plurality of other segments;
a discrepancy between a slope associated with the segment and slopes associated with a plurality of segments;
a discrepancy between a slope associated with the sub-segment and the slope associated with the segment;
a size of the segment; or
the slope associated with the sub-segment.
3. The computing device of claim 1 , wherein the plurality of data point pairs comprises substantially all combinations of data points within the plurality of data points.
4. The computing device of claim 1 , wherein each data point pair in the plurality of data point pairs comprises two subsequent data points.
5. The computing device of claim 1 , wherein the instructions, when executed by the processor, further cause the processor to determine whether the time series comprises the trend based on a standard normal variable calculated based on the stability significance of the time series.
6. The computing device of claim 1 , wherein the instructions, when executed by the processor, further cause the processor to calculate a weight for each data point pair in the plurality of data point pairs based on a first time associated with a first data point in the each data point pair and a second time associated with a second data point in the each data point pair.
7. The computing device of claim 1 , wherein the trend is one of a plurality of trends, the trend score is one of a plurality of trend scores corresponding to the plurality of trends, and the instructions, when executed by the processor, further cause the processor to provide data for display of a summary of the plurality of trends sorted in accordance with the plurality of trend scores.
8. A method comprising:
obtaining, via a processor, a time series comprising a plurality of data points associated with a sub-segment of a segment;
obtaining, via the processor, a plurality of weights associated with a plurality of data point pairs from the plurality of data points;
based on the plurality of weights and the plurality of data point pairs, determining, via the processor, whether the time series comprises a trend, wherein the determination whether the time series comprises the trend comprises determining a stability significance of the time series based on a weighted sum of a plurality of signs corresponding to differences of the plurality of data point pairs; and
based on a result of the determination of whether the time series comprises the trend:
calculating, via the processor, a trend score for the trend based on at least one characteristic of at least one of the segment and the sub-segment; and
providing, via the processor, data for display of the trend.
9. The method of claim 8 , further comprising calculating the trend score based on the stability significance of the time series and at least one of:
a discrepancy between the stability significance of the time series and stability significances of a plurality of other time series associated with the segment;
a discrepancy between a stability significance of the segment and stability significances of a plurality of other segments;
a discrepancy between a slope associated with the segment and slopes associated with a plurality of segments;
a discrepancy between a slope associated with the sub-segment and the slope associated with the segment;
a size of the segment; or
the slope associated with the sub-segment.
10. The method of claim 8 , wherein obtaining the plurality of weights comprises determining a weight for each data point pair in the plurality of data point pairs based on a first season associated with a first data point in the each data point pair and a second season associated with a second data point in the each data point pair.
11. The method of claim 8 , wherein obtaining the plurality of weights comprises determining a weight for each data point pair in the plurality of data point pairs based on at least one of a first time associated with a first data point in the each data point pair and a second time associated with a second data point in the each data point pair.
12. A non-transitory machine-readable storage medium storing instructions that, when executed by a processor of a computing device, cause the computing device to:
obtain a time series comprising a plurality of data points associated with a sub-segment of a segment;
determine a plurality of data point pairs in the plurality of data points;
determine a plurality of weights based on the plurality of data point pairs;
based on the plurality of data point pairs and the plurality of weights, determine whether the time series comprises a trend, wherein the determination of whether the time series comprises the trend comprises determining a stability significance of the time series based on a weighted sum of a plurality of signs corresponding to differences of the plurality of data point pairs; and
based on a result of the determination of whether the time series comprises the trend, calculate a trend score for the trend based on at least one characteristic of at least one of the segment or the sub-segment.
13. The non-transitory machine-readable storage medium of claim 12 , wherein the instructions, when executed by the processor, further cause the computing device to calculate the trend score based on the stability significance of the time series and at least one of:
a discrepancy between the stability significance of the time series and stability significances of plurality of other time series associated with the segment;
a discrepancy between a stability significance of the segment and stability significances of a plurality of other segments;
a discrepancy between a slope associated with the segment and slopes associated with a plurality of segments;
a discrepancy between a slope associated with the sub-segment and the slope associated with the segment;
a size of the segment; or
the slope associated with the sub-segment.
14. The non-transitory machine-readable storage medium of claim 12 , wherein the instructions, when executed by the processor, further cause the computing device to convert the time series to a relative time series and calculate the trend score based on the relative time series.
15. The method of claim 8 , wherein the determination whether the time series comprises the trend further comprises evaluating a standard normal variable calculated based on the stability significance of the time series.
16. The non-transitory machine-readable storage medium of claim 12 , wherein the instructions, when executed by the processor, further cause the computing device to determine whether the times series comprises the trend based on a standard normal variable calculated based on the stability significance of the time series.