Re: [R] Robust SE & Heteroskedasticity-consistent estimation

2010-05-18 Thread RATIARISON Eric
Thanks all of you. -Message d'origine- De : Arne Henningsen [mailto:arne.henning...@googlemail.com] Envoyé : mardi 18 mai 2010 06:00 À : RATIARISON Eric Cc : r-help@r-project.org; Ott-Siim Toomet; Achim Zeileis Objet : Re: [R] Robust SE & Heteroskedasticity-consistent estimati

Re: [R] Robust SE & Heteroskedasticity-consistent estimation

2010-05-17 Thread Arne Henningsen
; Should be corrected with possible use of sandwich should'nt it? I don't think that just using the sandwich method is sufficient to calculate consistent covariance matrices, when there are (equality or inequality) constraints on the parameters. /Arne > -Message d'origine-

Re: [R] Robust SE & Heteroskedasticity-consistent estimation

2010-05-17 Thread Arne Henningsen
i.e. in the BHHH form). b) The package with the new features is not yet available on CRAN but on R-Forge: http://r-forge.r-project.org/scm/?group_id=254 Please let me know if you experience any problems. /Arne > -Message d'origine- > De : Arne Henningsen [mailto:arne.henning...@

Re: [R] Robust SE & Heteroskedasticity-consistent estimation

2010-05-13 Thread Ott-Siim Toomet
stfun' applicable pour un objet de classe > "c('maxLik', 'maxim', 'list')" > > > > > > -Message d'origine- > De : Arne Henningsen [mailto:arne.henning...@googlemail.com] > Envoyé : mardi 11 mai 2010 08:25 > À : A

Re: [R] Robust SE & Heteroskedasticity-consistent estimation

2010-05-13 Thread RATIARISON Eric
pas de méthode pour 'estfun' applicable pour un objet de classe "c('maxLik', 'maxim', 'list')" -Message d'origine- De : Arne Henningsen [mailto:arne.henning...@googlemail.com] Envoyé : mardi 11 mai 2010 08:25 À : Achim Zeileis; RATIARISON Eric;

Re: [R] Robust SE & Heteroskedasticity-consistent estimation

2010-05-11 Thread RATIARISON Eric
x27;, 'list')"" -Message d'origine- De : RATIARISON Eric Envoyé : mardi 11 mai 2010 11:47 À : 'Arne Henningsen'; Achim Zeileis; r-help@r-project.org; Ott-Siim Toomet Objet : RE: [R] Robust SE & Heteroskedasticity-consistent estimation Thank you al

Re: [R] Robust SE & Heteroskedasticity-consistent estimation

2010-05-11 Thread Achim Zeileis
e "c('maxLik', 'maxim', 'list')"" -Message d'origine- De : RATIARISON Eric Envoyé : mardi 11 mai 2010 11:47 À : 'Arne Henningsen'; Achim Zeileis; r-help@r-project.org; Ott-Siim Toomet Objet : RE: [R] Robust SE & Heteroskedastic

Re: [R] Robust SE & Heteroskedasticity-consistent estimation

2010-05-11 Thread RATIARISON Eric
axlik,vcov.=vcovhc) do you think it's sound good ? -Message d'origine- De : Arne Henningsen [mailto:arne.henning...@googlemail.com] Envoyé : mardi 11 mai 2010 08:25 À : Achim Zeileis; RATIARISON Eric; r-help@r-project.org; Ott-Siim Toomet Objet : Re: [R] Robust SE & Heterosked

Re: [R] Robust SE & Heteroskedasticity-consistent estimation

2010-05-11 Thread Achim Zeileis
On Tue, 11 May 2010, Arne Henningsen wrote: On 11 May 2010 00:52, Achim Zeileis wrote: On Mon, 10 May 2010, RATIARISON Eric wrote: I'm using maxlik with functions specified (L, his gradient & hessian). Now I would like determine some robust standard errors of my estimators. So I 'm try to u

Re: [R] Robust SE & Heteroskedasticity-consistent estimation

2010-05-10 Thread Arne Henningsen
On 11 May 2010 00:52, Achim Zeileis wrote: > On Mon, 10 May 2010, RATIARISON Eric wrote: >> I'm using maxlik with functions specified (L, his gradient & hessian). >> >> Now I would like determine some robust standard errors of my estimators. >> >> So I 'm try to use vcovHC, or hccm or robcov for e

Re: [R] Robust SE & Heteroskedasticity-consistent estimation

2010-05-10 Thread Achim Zeileis
On Mon, 10 May 2010, RATIARISON Eric wrote: Hi, I'm using maxlik with functions specified (L, his gradient & hessian). Now I would like determine some robust standard errors of my estimators. So I 'm try to use vcovHC, or hccm or robcov for example but in use one of them with my result of ma

[R] Robust SE & Heteroskedasticity-consistent estimation

2010-05-10 Thread RATIARISON Eric
Hi, I'm using maxlik with functions specified (L, his gradient & hessian). Now I would like determine some robust standard errors of my estimators. So I 'm try to use vcovHC, or hccm or robcov for example but in use one of them with my result of maxlik, I've a the following error message :