Time Window Calculations

Time Window calculations allow new columns to be created that are based on a defined time window.

As a consequence there are several additional functions:

Functions	Description
COUNT_TIMEWIN	Like SUM_TIMEWIN, but simply returns the number of time slices between the defined time window, that have non-null values.
CUMSUM_TIMEWIN	The cumulative sum of Time Series value between start and end times. Alias for SUM_TIMEWIN(“Measure”,TimeWindowStart,Now).
LOOKUP	The value of a Time Series measure at a specific time.
MAX_TIMEWIN	The maximum value between the start and end times.
MEAN_TIMEWIN	The mean value of the Time Series between the start and end times.
MIN_TIMEWIN	The minimum value between the start and end times.
PRODUCT_TIMEWIN	The product of the Time Series values between the start and end times.
STDEV_TIMEWIN	The standard deviation of the time series between the start and end times.
STDEVP_TIMEWIN	The population standard deviation of the time series between the start and end times.
SUM_TIMEWIN	The sum of Time Series values between start and end times.

And three additional measures:

Measures	Description
SnapshotTime	The time slice at the Snapshot Time.
TimeWindowEnd	The Time slice at the end of a time window.
TimeWindowStart	The Time slice at the start of a time window.

When using Time Window calculations, fields referenced by the calculation should be enclosed in double quotes and NOT square brackets.

For example, using the Time Series column PRICE, the following calculations can be created:


Difference Between Start and End of Time Window	LOOKUP(“PRICE”,TimeWindowStart)-LOOKUP(“PRICE”,TimeWindowEnd)
Time Window Maximum value across	MAX_TIMEWIN( “PRICE”, TimeWindowStart, TimeWindowEnd )
Time Window Standard Deviation	STDEV_TIMEWIN( “PRICE”, TimeWindowStart, TimeWindowEnd )
Variance since Time Window Start	( [PRICE] - LOOKUP(“PRICE”,TimeWindowStart) ) / LOOKUP(“PRICE”,TimeWindowStart)