F. Porter, INTERVAL ESTIMATION USING THE LIKELIHOOD FUNCTION, Nuclear instruments & methods in physics research. Section A, Accelerators, spectrometers, detectors and associated equipment, 368(3), 1996, pp. 793-803
The general properties of two commonly-used methods of interval estima
tion for population parameters in physics are examined. Both of these
methods employ the likelihood function: (i) Obtaining an interval by f
inding the points where the likelihood decreases from its maximum by s
ome specified ratio; (ii) Obtaining an interval by finding points corr
esponding to some specified fraction of the total integral of the like
lihood function. In particular, the conditions for which these methods
give a confidence interval are illuminated, following an elaboration
on the definition of a confidence interval. The first method, in its g
eneral form, gives a confidence interval when the parameter is a funct
ion of a location parameter. The second method gives a confidence inte
rval when the parameter is a location parameter. A potential pitfall o
f performing a likelihood analysis without understanding the underlyin
g probability distribution is discussed using an example with a normal
likelihood function.