loss_fr¶
-
loss.loss_fr(sigma_hat, sigma)[source]¶ Squared Frobenius norm scaled by 1/p. Same as
np.linalg.norm(sigma_hat - sigma, 'fro')**2 *1/p.- Parameters
- sigma_hatnumpy.ndarray
The covariance matrix estimate using the estimator of interest.
- sigmanumpy.ndarray
The (true) population covariance matrix.
- Returns
- outfloat
The minimum variance loss.
Notes
The loss function is given by:
\[\mathcal{L}_{n}^{\mathrm{FR}}\left(\widehat{\Sigma}_{n}, \Sigma_{n}\right):=\frac{1}{p} \operatorname{Tr}\left[\left(\widehat{\Sigma}_{n} -\Sigma_{n}\right)^{2}\right]\]
