puryabaghaei at gmail.com
Mon Aug 18 17:45:14 EST 2008
We want to find the values of theta* *and* *delta* *is such a way that the
likelihood of the data is maximized. Imagine the curve of the log likelihood
function. The maxima of the curve is the highest point on the curve. We want
to find the values of theta* *and* *delta* *in such a way that the function
exactly indicates that point. Since at the maxima of a function the slope is
zero and the first derivative of a function is its slope, we differentiate
the function and set the derivative (slope) equal to zero. Having done this,
one gets values for theta and* *delta that maximize the likelihood of the
data, hence, maximum likelihood estimation*.*
On 8/18/08, Anthony James <luckyantonio2003 at yahoo.com> wrote:
> Dear all,
> I try to keep this as short as possible.
> I'm reading estimation procedure.
> There's a step where the log likelihood function is differentiate and set
> equal to 0. Why?
> Rasch mailing list
> Rasch at acer.edu.au
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Rasch