About the exceeded standard deviations type of the Trend component
The exceeded standard deviations (ESD) type of the Trend component compares the value in a current time period to the average of that value over a historical time period.
The comparison is made in terms of a specified number of standard deviations, with the standard deviation itself being based on the historical assemblage of values. An ESD trend detects activity that exceeds the specified number of (positive or negative) standard deviations over specific time period.
This type of Trend component allows the detection of a jump or drop that is significant not just in its size but also in comparison to how much the value has typically jumped or dropped for this entity in the past.
To detect an upward ESD spike, set the number of standard deviations value to a positive value. To detect a downward trend, set it to a negative value.
Example of an exceeded standard deviations component
The options for the exceeded standard deviations component
Copyright IBM Corporation 2012. All Rights Reserved.
