The above is glib in many ways---especially in its treatment of
errors. We will repair this later with real live theorems. First, we
state the obvious generalization.
The general maximum likelihood method is to put in Bayes Principle (*) and form for any measurement whose
probability depends on value and
theoretical parameter(s) .
The maximum likelihood estimate of is where
is a maximum. The error is
the standard deviation of the distribution of L as a function of
, i.e.,
in Bayes Principle (*) and form 
for any measurement 
whose
probability 
depends on value 
and
theoretical parameter(s) 
.
is 
where
is a maximum. The error 
is
the standard deviation of the distribution of L as a function of
, i.e.,
