[R-br] Mallows - estimador robusto

Benilton Carvalho beniltoncarvalho em gmail.com
Sexta Dezembro 14 09:07:24 BRST 2012


nada reproduzivel...

verifique se algumas de suas variaveis nao sao combinacoes lineares de
outras.




On 14 December 2012 10:52, Natália Bordin Barbieri <
nataliabordinbarbieri em gmail.com> wrote:

> Olá,
>
> preciso ajustar o seguinte modelo de estimação robusta, através do método
> de "Mallows' para o modelo de regressão logística que segue:
>
>
> require(robust)
> DM.glmRob3<-glmRob(DM ~ INFLAMACAO + IDADE + SEXOM + RACACOR + HIPERT +
> IMC + RCQ + TRIGT, family=binomial, data=DM, weights=NULL,
>         method ="mallows", model = TRUE, control = glmRob.control)
>
> Porém, ao rodar, o que retorna é o seguinte erro:
>
> 'The covariance matrix has become singular during
> the iterations of the MCD algorithm.
> There are 284 observations (in the entire dataset of 400 obs.) lying
> on the hyperplane with equation a_1*(x_i1 - m_1) + ... + a_p*(x_ip - m_p) =
> 0 with (m_1,...,m_p) the mean of these observations and coefficients a_i
> from the vector a <- c(1, 0, 0, 0, 0, 0, 0, 0) Erro em
> solve.default(V)  rotina Lapack dgesv: sistema é exatamente singular'
>
> Alguém já trabalhou com isso?Sabe me ajudar?
> Obrigada, Natália
>
> _______________________________________________
> R-br mailing list
> R-br em listas.c3sl.ufpr.br
> https://listas.inf.ufpr.br/cgi-bin/mailman/listinfo/r-br
> Leia o guia de postagem (http://www.leg.ufpr.br/r-br-guia) e forneça
> código mínimo reproduzível.
>
-------------- Próxima Parte ----------
Um anexo em HTML foi limpo...
URL: <http://listas.inf.ufpr.br/pipermail/r-br/attachments/20121214/13db214e/attachment.html>


Mais detalhes sobre a lista de discussão R-br