On Robustness of Principal Component Regression: Author Response