Regress on matrix on vector to obtain rank-one approximation in Partial Least Square

16 Views Asked by At

I'm trying to understand the equation for partial least squares (PLS) regression from the paper "A survey of partial least squares (PLS) methods, with emphasis on the two-blocks case". I'm having trouble understanding how the equation is derived based on the paper. Can someone please explain it to me? Here is a more detailed explanation of the question:

The equation for PLS regression is given in the paper as follows:

$$ \hat{X}_1(\epsilon_1) = \epsilon_1((\epsilon_1^T \epsilon_1)^{-1} \epsilon_1^T X_1) $$

This paper tell me that we want to regress