Doubly reweighted estimators for the parameters of the multivariate t-distribution

Estimação robusta para a distribuição Skew-t

“…para todos os parâmetros da distribuição Skew-t. Fornecemos um algoritmo EM modificado para o cálculo dos estimadores ML q , mostrando a robustez do método e um melhor ajuste do modelo ST às observações atípicas, ao se estimar de forma simultânea os graus de liberdade com os demais parâmetros do modelo de interesse. Os resultados obtidos neste trabalho são uma generalização do trabalho de [13] que estudou o método ML q para a distribuição t-Student. A aplicação numérica foi realizada no programa R; veja [14] para mais detalhes.…”

Robust Stochastic Gradient Descent With Student-t Distribution Based First-Order Momentum

“…In order for the EMA-based momentum to be robust, the distribution of the gradients must be assumed to come from a robust probability distribution that can yield a robust mean estimator. We, therefore, propose to replace the normal distribution momentum estimator with one drawn from the student-t distribution, which is well-known to be a robust probability distribution [10], [29], [30], as shown in Fig. 1, and a general form of the normal distribution.…”

Multivariate Student versus Multivariate Gaussian Regression Models with Application to Finance

“…They used the framework of normal mixture distributions in which the Student distribution can be expressed as a combination of a Gaussian random variable and an inverse gamma random variable. More recently, the work in Dogru et al (2018) proposed a more robust extension, replacing maximum likelihood by a kind of M-estimation method based on the minimization of a q-entropy criterion. For the multivariate Student IT model, the work in Prucha and Kelejian (1984) derived the normal equations for the maximum likelihood estimators and their asymptotic properties with known degrees of freedom in a framework that encompasses our multivariate Student regression case.…”