Page 540 - Probability and Statistical Inference
P. 540
11. Likehood Ratio and Other Tests 517
Now, we combine (11.3.5)-(11.3.6) to express the likelihood ratio as
with from (11.3.4) and reject H if and only if ? is small. Thus, we decide as
0
follows:
where k(> 0) is a generic constant. Again, let us utilize (11.2.4) and write
which implies that
In other words, the small values of
will correspond to the large values of
Thus, we can rewrite the test (11.3.8) as follows:
where k(> 0) is an appropriate number.
Figure 11.3.1. Two-Sided Students t Rejection Region
v
Where the Degree of Freedom Is v = n + n - 2
1 2

