at work

How Likelihood* Works

 * Likelihood is the probability of the data.

There are more complicated definitions, like

The density function of continuous random variables, X = X1, X2, ..., Xn, that depend on model parameters, q = q1, q2, ..., qk, and written as f(X | q), is the density function of X, given q, or is the likelihood of q, given X, often written as L(q | X).      (Which definition do you prefer?)


In the literature L can also represent log(likelihood).


Many years ago, before R was available and statistical procedures needed to be programmed in FORTRAN, I discussed likelihood here.

... I discussed generalized linear models here.


SElogo.gif (5955 bytes)

[ Home ]   [ Feedback ]

Mail to
Copyright � 1998-2008 Charles Annis, P.E.
Last modified: December 16, 2009