This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Second-order inclusion probability: the probability, denoted ij , that a given pair of units (say i and j ) will appear in the sample, under a given design. SRS: ij n ( n 1) / ( N ( N 1)) , 1 i 6 = j N . Population total: := N i =1 x i . Population mean: := 1 N N i =1 x i . Popula-tion variance: 2 := 1 N 1 N i =1 ( x i ) 2 . Horvitz-Thompson estimator (HTE) of the population total: = i sample x i i . HTE of mean: = 1 N i sample x i i . SRS: = N x ; = x . HTE is unbiased under any design: E ( ) = . 1 SRS: Var ( N x ) = N 2 (1 n/N ) 2 n . Finite population correction (fpc): 1 n/N . E ( s 2 ) = 2 . SE ( N x ) = N q 1 n/N s n . SE ( x ) = q 1 n/N s n . 2...
View Full Document
- Spring '11