Reading for Today's Lecture:
Goals of Today's Lecture:
Binomial(n,p): log likelihood
(part depending on
)
is function of X alone, not
of
as well.
Normal example:
is, ignoring terms not containing
,
Examples of the Factorization Criterion:
Theorem: If the model for data X has density
then the statistic S(X) is sufficient if and only if the density can be
factored as
Proof: Find statistic T(X) such that
X is a one to one function of the pair S,T. Apply
change of variables to the joint density of S and T. If the density
factors then
Conversely if S is sufficient
then the fT|S has no
in it so joint
density of S,T is
Example: If
are iid
then the joint density is
which is evidently a function of
Definition: A statistic T is complete for a model
if
We have already seen that X is complete in the Binomial(n,p) model.
In the
model suppose
There is only one general tactic. Suppose X has
density
You prove the sufficiency by the factorization criterion and
the completeness using the properties of Laplace transforms
and the fact that the joint density of
Example:
model density
has form
Remark: The statistic
is a one to one
function of
so it must be complete
and sufficient, too. Any function of the latter statistic
can be rewritten as a function of the former and vice versa.
Theorem: If S is a complete sufficient
statistic for some model and h(S) is an unbiased
estimate of some parameter
then h(S) is the
UMVUE of
.
Proof: Suppose T is another unbiased estimate
of .
According to Rao-Blackwell, T is improved
by E(T|S) so if h(S) is not UMVUE then there must exist
another function h*(S) which is unbiased and whose variance
is smaller than that of h(S) for some value of
.
But
Example: In the
example the random
variable
has a
distribution.
It follows that
Binomial(n,p) log odds is
.
Since
the expectation of any function of the data is a polynomial function
of p and since
is not a polynomial function of p there
is no unbiased estimate of
In any model
is sufficient. In any iid
model the vector
of order statistics
is
sufficient. In
model we have 3 sufficient
statistics:
Notice that I can calculate S3 from the values of S1 or S2but not vice versa and that I can calculate S2 from S1 but not vice
versa. It turns out that
is a minimal sufficient
statistic meaning that it is a function of any other sufficient statistic.
(You can't collapse the data set any more without losing information about
.)
Recognize minimal sufficient statistics from :
Fact: If you fix some particular
then
the log likelihood ratio function
Subtraction of
gets rid of irrelevant constants
in
.
In
example:
FACT: A complete sufficient statistic is also minimal sufficient.