Hints for problem 3.16

Note that when the predictors are orthonormal, then $\boldsymbol{X}^T \boldsymbol{X} = I$ and the LS estimate of $\beta$ is given by $\hat{\beta} = (\boldsymbol{X}^T \boldsymbol{X})^{-1} \boldsymbol{X} \boldsymbol{y} = \boldsymbol{X} \boldsymbol{y}$.

Published Aug. 8, 2025 8:03 AM - Last modified Aug. 8, 2025 8:03 AM