See
Brian Mamly- Multivariate Statistical Methods : A Primer
is a very nice exposition of your problem
Basilio de Bragança Pereira
Federal University of Rio de Janeiro
Mary Swinson Escreveu:
> Hi All,
>
> I have 2 sets of multivariate measurements that I would like to compare with some kind of hypothesis test (multivariate t-test?). Any thoughts/pointers/references would be most welcome. Here is an example I want to compare Method 1 vs. Method 2
>
> Method 1 has N_1 multivariate observations {e1_1,....e1_N1} where e1_i={e1_1i, e1_2i, ..., e1_pi} and p is the number of features in the ith observation (p is fixed)
>
> Method 2 has N_2 multivariate observations {e2_1,....e2_N2} where e2_i={e2_1i, e2_2i, ..., e2_pi} and p is the number of features in the ith observation (p is fixed)
>
> I want to test the hypothesis that Method 1 = Method 2, using {e1_1, e1_2, ..., e1_N1} and {e_21, e2_2, ..., e2_N2}, a sort of multivariate equivalent of the t-test?
>
> Best Regards,
>
> Mary
>
>
> ---------------------------------
> Sponsored Link
>
> Rates near 39yr lows. $420,000 Loan for $1399/mo - Calculate new house payment
|