Reading for Today's Lecture:
Goals of Today's Lecture:
Today's notes
A level
confidence set for a parameter
is a random subset C, of the set of possible values of
such that for each
we have
Suppose C is a level
confidence set for
.
To test
we consider the test which rejects if
.
This test has level
.
Conversely, suppose
that for each
we have available a level
test
of
who rejection region is say
.
Then if
we define
we get
a level
confidence for
.
The usual t test gives
rise in this way to the usual t confidence intervals
Definition: A pivot (or pivotal quantity) is a function
whose distribution is the same for all
.
(As usual
the
in the pivot is the same
as the one being used to
calculate the distribution of
.
Pivots can be used to generate confidence sets as follows. Pick a set
A in the space of possible values for g. Let
;
since g is pivotal
is the same for all
.
Now given
a data set X solve the relation
Example: The quantity
In the same model we also have
In general the interval from
to
has level
.
For a fixed value of
we can
minimize the length of the resulting interval numerically. This sort of optimization
is rarely used. See your homework for an example of the method.