Re: [R] maxNR in maxLik package never stops

2010-08-30 Thread Ott-Siim Toomet
Hi, sure, maxNR uses 'iterlim' (or was it 'maxIter'?) where you can specify the max # of iterations. The default was 100 iterations, AFAIK. There was a bug in one of the previous versions which lead to infinite loops of constrained maximization. Do you use constrained or unconstrained? and

Re: [R] Robust SE Heteroskedasticity-consistent estimation

2010-05-13 Thread Ott-Siim Toomet
- 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 Heteroskedasticity-consistent estimation On 11 May 2010 00:52, Achim Zeileis achim.zeil

Re: [R] Interpreting coefficient in selection and outcome Heckman models in sampleSelection

2010-01-04 Thread Ott-Siim Toomet
Hi Mark, why do you need that? If your task is to estimate how much your y changes if x change, why not use simple OLS? (Well, right, you should be able to use sampleSelection as well). It shouldn't probably be hard to compute it -- it is just OLS marginal effect + som kind of derivative of