Suppose that the loss function is the piecewise linear function [ operatorname{Loss}(y, hat{y})=alpha(hat{y}-y)_{+}+beta(y-hat{y})_{+}, quad alpha, beta>0 ]
Question:
Suppose that the loss function is the piecewise linear function
\[ \operatorname{Loss}(y, \hat{y})=\alpha(\hat{y}-y)_{+}+\beta(y-\hat{y})_{+}, \quad \alpha, \beta>0 \]
where \(c_{+}\)is equal to \(c\) if \(c>0\), and zero otherwise. Show that the minimizer of the risk \(\ell(g)=\mathbb{E} \operatorname{Loss}(Y, g(\boldsymbol{X}))\) satisfies
\[ \mathbb{P}\left[Y In other words, \(g^{*}(\boldsymbol{x})\) is the \(\beta /(\alpha+\beta)\) quantile of \(Y\), conditional on \(\boldsymbol{X}=\boldsymbol{x}\).
Fantastic news! We've Found the answer you've been seeking!
Step by Step Answer:
Related Book For
Data Science And Machine Learning Mathematical And Statistical Methods
ISBN: 9781118710852
1st Edition
Authors: Dirk P. Kroese, Thomas Taimre, Radislav Vaisman, Zdravko Botev
Question Posted: