Helen:
It is good practice to reply not only to the responder to your post, but also to
the list in general. I don't know of any methods for non-parametric ability
estimation. The irt.ability function provides you with three options for
estimating ability: the maximum likelihood estimate (MLE), the maximum a
posteriori estimate (MAP), and the expected a posteriori (EAP).
Both the MAP and the EAP assume a normal prior with N(mu, sigma^2). Now, the MLE
uses no such assumption, so it is distribution-free in the sense that there is
no assumption about a distributional shape, but MLEs don't classify as a
nonparametric method per se.
From: Helen Lisman [mailto:hlnlisman@gmail.com]
Sent: Wednesday, March 10, 2010 4:29 PM
To: Doran, Harold
Subject: Re: [R] help R non-parametric IRT simulation
What about ability parameter simulation in non-parametric method? is irt.ability
has any subfunction to do non-parametric simulation for ability estimate?
thank you much--
On Wed, Mar 10, 2010 at 2:12 PM, Doran, Harold
<HDoran@air.org<mailto:HDoran@air.org>> wrote:
There is no function in R AFAIK, where an IRT model is estimated using the NPMML
method (non-parametric marginal maximum likelihood). All MML methods in R assume
the population distribution is normal.
Now, the jml function in the MiscPsycho package uses joint maximum likelihood
for estimating item parameters for the Rasch model. Now, jml makes no particular
assumptions about the shape of the proficiency distribution or the shape of the
item distribution.
However, jml estimates are not consistent, but they tend to the true parameter
with less bias as the number of items increases. So, it may be useful for you in
some regard.
-----Original Message-----
From: r-help-bounces@r-project.org<mailto:r-help-bounces@r-project.org>
[mailto:r-help-bounces@r-project.org<mailto:r-help-bounces@r-project.org>]
On Behalf Of Helen Lisman
Sent: Wednesday, March 10, 2010 3:03 PM
To: r-help@r-project.org<mailto:r-help@r-project.org>
Subject: Re: [R] help R non-parametric IRT simulation
Hello R,
I am looking for non-parametric simulation in IRT. Is there any IRT
package that does non-parametric simulation?
helen L
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org<mailto:R-help@r-project.org> mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html<http://www.r-project.org/posting-guide.html>
and provide commented, minimal, self-contained, reproducible code.
[[alternative HTML version deleted]]