A New Concept for Tuning Design Weights in Survey Sampling: Jackknifing in Theory and Practice introduces the new concept of tuning design weights in survey sampling by presenting three concepts: calibration, jackknifing, and imputing where needed. This new methodology allows survey statisticians to develop statistical software for analyzing data in a more precisely and friendly way than with existing techniques.Explains how to calibrate design weights in survey samplingDiscusses how Jackknifing is needed in design weights in survey samplingDescribes how design weights are imputed in survey sampling
Muu info
This book introduces a new statistical survey methodology that can be used to estimate population parameters, presenting three concepts for discussion, including calibration, jackknifing, and imputing where needed.
| Preface |
|
xi | |
|
|
|
1 | (26) |
|
|
|
1 | (1) |
|
1.2 Estimation problem and notation |
|
|
1 | (4) |
|
1.3 Modeling of jumbo pumpkins |
|
|
5 | (3) |
|
1.4 The concept of jackknifing |
|
|
8 | (3) |
|
1.5 Jackknifing the sample mean |
|
|
11 | (3) |
|
1.6 Doubly jackknifed sample mean |
|
|
14 | (2) |
|
1.7 Jackknifing a sample proportion |
|
|
16 | (1) |
|
1.8 Jackknifing of a double suffix variable sum |
|
|
17 | (1) |
|
1.9 Frequently asked questions |
|
|
17 | (1) |
|
|
|
18 | (9) |
|
2 Tuning of jackknife estimator |
|
|
27 | (36) |
|
|
|
27 | (1) |
|
|
|
27 | (1) |
|
2.3 Tuning with a chi-square type distance function |
|
|
28 | (13) |
|
2.4 Tuning with dell function |
|
|
41 | (7) |
|
|
|
48 | (1) |
|
|
|
48 | (15) |
|
3 Model assisted tuning of estimators |
|
|
63 | (22) |
|
|
|
63 | (1) |
|
3.2 Model assisted tuning with a chi-square distance function |
|
|
63 | (6) |
|
3.3 Model assisted tuning with a dual-to-empirical log-likelihood (dell) function |
|
|
69 | (5) |
|
|
|
74 | (11) |
|
4 Tuned estimators of finite population variance |
|
|
85 | (54) |
|
|
|
85 | (1) |
|
4.2 Tuned estimator of finite population variance |
|
|
85 | (3) |
|
4.3 Tuning with a chi-square distance |
|
|
88 | (10) |
|
4.4 Tuning of estimator of finite population variance with a dual-to-empirical log-likelihood (dell) function |
|
|
98 | (9) |
|
4.5 Alternative tuning with a chi-square distance |
|
|
107 | (6) |
|
4.6 Alternative tuning with a dell function |
|
|
113 | (5) |
|
|
|
118 | (21) |
|
5 Tuned estimators of correlation coefficient |
|
|
139 | (26) |
|
|
|
139 | (1) |
|
5.2 Correlation coefficient |
|
|
139 | (1) |
|
5.3 Tuned estimator of correlation coefficient |
|
|
140 | (11) |
|
|
|
151 | (14) |
|
6 Tuning of multicharacter survey estimators |
|
|
165 | (34) |
|
|
|
165 | (1) |
|
6.2 Transformation on selection probabilities |
|
|
165 | (1) |
|
6.3 Tuning with a chi-square distance function |
|
|
166 | (13) |
|
6.4 Tuning of the multicharacter estimator of population total with dual-to-empirical log-likelihood function |
|
|
179 | (12) |
|
|
|
191 | (8) |
|
7 Tuning of the Horvitz-Thompson estimator |
|
|
199 | (20) |
|
|
|
199 | (1) |
|
7.2 Jackknifed weights in the Horvitz-Thompson estimator |
|
|
199 | (1) |
|
7.3 Tuning with a chi-square distance function while using jackknifed sample means |
|
|
200 | (8) |
|
7.4 Tuning of the Horvitz-Thompson estimator with a displacement function |
|
|
208 | (6) |
|
|
|
214 | (5) |
|
8 Tuning in stratified sampling |
|
|
219 | (38) |
|
|
|
219 | (1) |
|
|
|
219 | (2) |
|
8.3 Tuning with a chi-square distance function using stratum-level known population means of an auxiliary variable |
|
|
221 | (11) |
|
8.4 Tuning with dual-to-empirical log-likelihood function using stratum-level known population means of an auxiliary variable |
|
|
232 | (10) |
|
|
|
242 | (15) |
|
9 Tuning using multiauxiliary information |
|
|
257 | (26) |
|
|
|
257 | (1) |
|
|
|
257 | (1) |
|
9.3 Tuning with a chi-square distance function |
|
|
258 | (9) |
|
9.4 Tuning with empirical log-likelihood function |
|
|
267 | (8) |
|
|
|
275 | (8) |
|
10 A brief review of related work |
|
|
283 | (8) |
|
|
|
283 | (1) |
|
|
|
283 | (6) |
|
|
|
289 | (2) |
| Bibliography |
|
291 | (6) |
| Author Index |
|
297 | (2) |
| Subject Index |
|
299 | |
Sarjinder Singh has a Ph.D. degree in statistics specializing in the field of survey sampling. Associate professor of mathematics and statistics, Texas A&M University Kingsville (h index 11). He is a founder of higher order calibration technique in survey sampling. His first paper on this topic was published in the journal Survey Methodology, Statistics Canada, during 1998. Later he published numerous papers on calibration technique, and this monograph is also based on calibration techniques but with a different aspect. He is also pioneer founder of a dual problem of calibration published in highly respectable journal Statistics-A Journal of Theoretical and Applied Statistics. He also introduced the pioneering idea of calibration using displacement function and published in an prestigious journal, Metrika. He has published over 150 research papers in the field of survey sampling. Stephen A. Sedory has a Ph.D. degree in Mathematics, and has over 20 years of teaching and research experience at graduate and undergraduate level (Associate Professor of Mathematics, Department of Mathematics, Texas A&M University-Kingsville. Although his previous work is in the field of Topology, he has recently been working in the field of survey sampling. He has introduced the idea of two-step calibration and calibrated maximum likelihood calibration weights jointly with the first author. Maria Del Mar Rueda is a full-Professor and Director of a research group focusing on design and analysis of sample surveys at the University of Granada, Spain. Antonio Arcos is an Assistant Professor of Statistics, University of Granada, Spain, and is also working in the same areas of survey sampling. Together with Maria, Antonio is not only an expert in survey sampling, but also in writing codes in R language. All R-codes in this monographs are written by Maria and Antonio. In addition, both have contributed several papers on the calibration technique in survey sampling. Raghunath Arnab has a Ph.D. in statistics with specialization in survey sampling from the Indian Statistical Institute. He is based at the Dept of Statistics, University of Botswana. He has published very good quality papers in the field of complex survey sampling. His major contribution in this monograph is to check all the theoretical derivations of the results.
Lisainfo e-raamatute kohta