In this paper, a Constrained Mismatched Maximum Likelihood (CMML) estimator for the joint estimation of the scatter matrix and the power of Complex Elliptically Symmetric (CES) distributed vectors is derived under misspecified data models. Specifically, this estimator is obtained by assuming a Normal model while the data are sampled from a complex tdistribution. The convergence point of such CMML estimator is investigated and its Mean Square Error (MSE) compared with the Constrained Misspecified Cramér-Rao Bound (CMCRB).
On scatter matrix estimation in the presence of unknown extra parameters: Mismatched scenario
FORTUNATI, STEFANO;GINI, FULVIO;GRECO, MARIA
2016-01-01
Abstract
In this paper, a Constrained Mismatched Maximum Likelihood (CMML) estimator for the joint estimation of the scatter matrix and the power of Complex Elliptically Symmetric (CES) distributed vectors is derived under misspecified data models. Specifically, this estimator is obtained by assuming a Normal model while the data are sampled from a complex tdistribution. The convergence point of such CMML estimator is investigated and its Mean Square Error (MSE) compared with the Constrained Misspecified Cramér-Rao Bound (CMCRB).File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
