Discriminant informant

A term in discriminant analysis denoting a variable used to establish a rule for assigning an object with measurements , drawn from a mixture of sets with distribution densities and a priori probabilities , to one of these sets. The -th discriminant informant of the object with measurement is defined as

where is the loss due to assigning an element from the distribution to the distribution . The rule for assigning an object to the distribution with the largest discriminant informant has minimum mathematical expectation of the loss. In particular, if all distributions are normal and have identical covariance matrices, all discriminant informants are linear. Then, if , the difference is Fisher's linear discriminant function.

References

[1]	C.R. Rao, "Linear statistical inference and its applications" , Wiley (1965)