Closed b-steve closed 10 years ago
When se.correct() is used on a model fit where g0 is 1, then the bias correction procedure will push the estimate outside the parameter space.
se.correct()
g0
No longer "correct" parameter values, so this is no longer relevant.
When
se.correct()
is used on a model fit whereg0
is 1, then the bias correction procedure will push the estimate outside the parameter space.