If we include to Linear Discriminant Analysis quadratic parameters we get Quadratic Discriminant Analysis classifier. Number of parameters is $(K-1)\times[(p\times(p+3)/2) + 1]$ where K is number of classes and p is number of features.

Why $K-1$? In the book it is said that we only need to compare $\delta_{k}(x) - \delta_{K}(x)$ of discriminant functions of some prespecified class K, but why do we do that and how it works?

$(p\times(p+3)/2) + 1$ is number of factors in multivariate quadratic function how is it calculated?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.