Abstract
The measures of information theory, suitably modified, can be used as measures of the strength of intervariable relations. This paper defines the normalized transmission between two variables; it is a measure of the degree to which knowledge of one variable is helpful in guessing the value of the other, and it can be applied to static or dynamic relations. The usefulness of the measure is illustrated by an example consisting of seven weather variables, where it is shown that the normalized transmission leads to a natural partition of the system into subsystems, within which the intervariable relations are strong and between which they are weak. Normalized transmission is also compared with other measures of relatedness.

This publication has 1 reference indexed in Scilit: