The DISTRIB operator contains a distribution on a variable. SENSOR allows the comparison of a mask with the current distribution in the DISTRIB, after the valid range and probability is taken into account.
The format is
SENSOR(Variable, Initiate)
where:
Variable is the variable having the distribution. An error will be returned if a distribution (DISTRIB filled with data) is not found.
Initiate controls the timing of the calculation, usually after analysis has completed.
In operation, a change in range (or distribution) of Variable1 will trigger a matching operation between the distribution in the DISTRIB and the mask in the SENSOR operator. A correlation value will be output as the value of the SENSOR function.
The purpose of the operator is to allow analysis on the basis of distributions to proceed, and then assess the result of the analysis in terms of a particular shape of the resulting distribution.
As with any feature detection, the calculation is expensive, so must be restricted in use, but it allows for subtle effects. The easiest way to minimise wasted effort is to enable the operator only after analysis of distributions has ceased.
The SENSOR operator is usually built and its mask defined using PieceLine.