Given there are two classes A and B and the prior probability of belonging to $ A = Na/N $ and $B = Nb/N $, I want to show that the linear discriminant analysis rule classifies an observation x to class B if:

$ x^TΣ(μ_{B} −μ_{A})> \frac{1}2μ^T_{B}Σ^{-1}μ_{B}−\frac{1}2μ^T_{A}Σμ_{A}+log(N_{A}/N) −log(N_{B}/N) $

I'm very new to this platform so I am unsure about how to code my math into this post! What I've done is attached my written attempt in a photo.

Is my line of thinking correct?

Any help at all would be greatly appreciated. Thank you!

enter image description here


Yes, your reasoning is correct.

By the way, you can just use the same Latex syntax to write your math as you've done with the classification rule.


Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.