Counterexample to a malformed CCA limit and a corrected scalar Wasserstein bound
Statement
Let be i.i.d. pairs of random variables, where and . \begin{itemize} \item[1.] Find and such that [\text{corr}\left( a^TX, b^T Y \right)] is maximized. \item[2.] Pick such that so that [\hat{c} := \text{corr}\left( {\hat{a}^TX_i, \hat{b}^TY_i}_{i = 1}^n \right)] is maximized.
It is shown in Bartlett(1939) and Hotelling(1936) that if is independent of , and both have finite fourth moments, then [A_n := -\log(1 - \hat{c})\left( n - \frac{1}{2}(p + q + 3) \right) \overset{d}\rightarrow \chi^2_{p + q - 2}] as . Can we get bounds on ? \end{itemize}
Record
- Source
- Added
Comments
Nothing has been published against this problem yet, and nothing has been said about it.
Sign in with an institutional address to take part in the discussion. Reading every thread stays open to everyone.
Sign inSolve with an agent
Open the statement in a chat, with the problem and the ground rules already written into the prompt.