Sie sind auf Seite 1von 7

Simple versus Composite Hypothesis

Recall, a simple hypothesis completely specifies the distribution. A


composite does not.
When testing a simple null hypothesis versus a composite
alternative, the power of the test is a function of the parameter of
interest.
In addition, the power is also affected by the sample size.

week 9

Example

week 9

Test for Mean of Normal Population 2 is known


Suppose X1, , Xn is a random sample from a N(, 2) distribution
where 2 is known. We are interested in testing hypotheses about .
The test statistics is the standardized version of the sample mean X .
We could test three sets of hypotheses

week 9

Test for Mean of Normal Population 2 is unknown


Suppose X1, , Xn is a random sample from a N(, 2) distribution
where 2 is unknown, n is small and we are interested in testing
hypotheses about .
The test statistics is...

week 9

Example
In a metropolitan area, the concentration of cadmium (Cd) in leaf
lettuce was measured in 6 representative gardens where sewage
sludge was used as fertilizer. The following measurements (in
mg/kg of dry weight) were obtained.
Cd: 21 38 12 15 14
8
Is there evidence that the mean concentration of Cd is higher than 12.

week 9

Test for Mean of a Non-Normal Population


Suppose X1, , Xn are iid from some distribution with E(Xi)= and
Var(Xi)= 2. Further suppose that n is large and we are interested in
testing hypotheses about .
Since n is large the CLT applies to the sample mean and the test
statistics is again the standardized version of the sample mean X .

week 9

Example Binomial Distribution


Suppose X1,,Xn are random sample from Bernoulli() distribution.
We are interested in testing hypotheses about

week 9

Das könnte Ihnen auch gefallen