{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

Handout3-W09

# Handout3-W09 - STAT 454/854 Winter 2009 Handout#3...

This preview shows pages 1–2. Sign up to view the full content.

STAT 454/854 Winter 2009 Handout #3 Single-stage Cluster Sampling (1) Single-stage cluster sampling with clusters of equal sizes We have M i = M , i = 1 , · · · , N and K = NM . The population mean is μ y = N - 1 N i =1 μ i , where μ i = M - 1 M j =1 y ij . Under single-stage cluster sampling with clus- ters selected by SRSWOR, an unbiased estimator of μ y is given by ¯ y = n - 1 n i =1 μ i , with theoretical variance V y ) = 1 - n N σ 2 M n , where σ 2 M = ( N - 1) - 1 N i =1 ( μ i - μ y ) 2 . An unbiased variance estimator is given by v y ) = 1 - n N s 2 M n , where s 2 M = ( n - 1) - 1 i s ( μ i - ¯ y ) 2 . (2) Intra-cluster correlation coefficient Suppose that we randomly select a cluster from the population, and then randomly select two elements without replacement in the selected cluster, and let z 1 and z 2 be the respective y -value of the two selected elements. Define the intra-cluster correlation coefficient as ρ = Cov ( z 1 , z 2 ) / { V ( z 1 ) V ( z 2 ) } 1 / 2 , where Cov ( z 1 , z 2 ) = 1 NM ( M - 1) N X i =1 M X j =1 M X k 6 = j ( y ij - μ y )( y ik - μ y ) and V ( z 1 ) = V ( z 2 ) = { ( NM -

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}