This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: March 5 th ( Large Sample ) Confidence interval for a population proportion p Let . . . ~ ( ), 1, , i i d i X Bernoulli p i n = L , please find the 100(1α)% CI for p. 1) Point estimator : 1 ˆ n i i X p X n = = = ∑ (ex. 1000 n = , ˆ 0.6 p = ) 2) 100(1α)% C.I. for p <Thm> Central Limit Theorem ( ) (0,1) ( ) n X E X N Var X → ∞ → When n is large enough, we have ( ) ~ (0,1) ( ) X E X Z N Var X = & Application #1. Inference on μ when the population distribution is unknown but the sample size is large ~ (0,1) / X Z N n μ σ = & ~ (0,1) / X Z N S n μ = & (Slutsky’s Theorem) => 100(1 )% α C.I. for μ /2 S X Z n α ± Application #2. Inference on one population proportion p when the population is Bernoulli(p) *** ( ) ~ (0,1) ( ) X E X Z N Var X = & 1 1 ( ) ( ) ( ) i i X E X E E X np p n n n = = = ⋅ = ∑ ∑ , ( ~ ( , )) i X B n p ∑ Q 2 2 1 1 (1 ) ( ) ( ) ( ) (1 ) i i X p p Var X Var Var X np p n n n n = = = = ∑ ∑ 1 ~ (0,1) (1 ) X p Z N p p n = & ˆ ~ (0,1)...
View
Full Document
 Spring '08
 Zhu,W

Click to edit the document details