9.16 MAXIMUM LIKELIHOOD ESTIMATION
Consider estimating parameter θ where its prior distribution 
is not known and cannot be used in MAP estimation. For this condition, we can develop an estimator similar to the MAP estimator which does not require any information about the prior distribution.
Definition: Likelihood Function The likelihood function associated with N random variables 
is the conditional pdf
(9.257) 
The log-likelihood function is
(9.258) 
where 
usually is the natural logarithm.
The likelihood function is a function of the unknown parameter θ for specific outcomes of the samples 
. We are interested in the value of θ for which 
is maximum; it is the “most likely” value for particular outcomes. The log-likelihood function is usually preferred because the logarithm often simplifies ...
Get Probability, Random Variables, and Random Processes: Theory and Signal Processing Applications now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
