8_more_ratio_estimation

8_more_ratio_estimation - H domains(categories 2 Estimate...

Info iconThis preview shows pages 1–8. Sign up to view the full content.

View Full Document Right Arrow Icon
More Ratio and Regression Estimation Lecture 8 STAT 651: Survey Sampling Methods, Kaizar – p.1/8
Background image of page 1

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

View Full DocumentRight Arrow Icon
Lecture 8: More Ratio Estimation Reading: Lohr Chapter 4, sections 4-5 Poststratification Stratified Ratio Estimation STAT 651: Survey Sampling Methods, Kaizar – p.2/8
Background image of page 2
Recall Stratification Recall Stratified Random Sampling: ¯ y str = H s h =1 N h N ¯ y h ˆ t str = H s h =1 ˆ t h = H s h =1 N h ¯ y h STAT 651: Survey Sampling Methods, Kaizar – p.3/8
Background image of page 3

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

View Full DocumentRight Arrow Icon
Poststratification Ratio Estimation is the process of adjusting estimates for known population means. Poststratification is the process of adjusting estimates for known population category sizes, or known domain sizes. STAT 651: Survey Sampling Methods, Kaizar – p.4/8
Background image of page 4
Poststratified Mean 1. Group the sample into H domains (categories) 2. Estimate the mean within each domain: ¯ y h , h = 1 , . . . , H 3. Take a weighted average of the domain means ¯ y post = H s h =1 N h N ¯ y h STAT 651: Survey Sampling Methods, Kaizar – p.5/8
Background image of page 5

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

View Full DocumentRight Arrow Icon
Poststratified Total 1. Group the sample into
Background image of page 6
Background image of page 7

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

View Full DocumentRight Arrow Icon
Background image of page 8
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: H domains (categories) 2. Estimate the mean within each domain: ¯ y h , h = 1 , . . . , H 3. Take a weighted sum of the domain means ˆ t ypost = H s h =1 N h ¯ y h STAT 651: Survey Sampling Methods, Kaizar – p.6/8 Variance Estimation Recall Domain Variance (assuming the sample size in the domain is large enough): V [¯ y h ] ≈ p 1-n N P s 2 yd n d We approximate the variance for the poststratified estimator using an independence approximation: V [¯ y str ] = V b H s h =1 N h N ¯ y h B = STAT 651: Survey Sampling Methods, Kaizar – p.7/8 Ratio Estimation in Stratification Two approaches: 1. Combined Ratio Estimator: ˆ t yrc = ˆ Bt x , where ˆ B = ˆ t ystr ˆ t xstr 2. Separate Ratio Estimator: ˆ t yrs = H s h =1 ˆ B h t xh , where ˆ B h = ˆ t yh ˆ t xh STAT 651: Survey Sampling Methods, Kaizar – p.8/8...
View Full Document

This note was uploaded on 07/26/2011 for the course STA 651 taught by Professor Kaizar during the Winter '11 term at Ohio State.

Page1 / 8

8_more_ratio_estimation - H domains(categories 2 Estimate...

This preview shows document pages 1 - 8. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online