New Variance Estimators in the Presence of Nonresponse Under Unequal Probability Sampling, with Application to Fine Particulate Matter in Thailand
Keywords:
Generalized regression estimator, joint inclusion probability, ratio estimator, response probability, variance estimatorAbstract
Fine particulate matter exacerbates the environmental problem of air pollution in Thailand and contributes to premature deaths. Monitoring of levels of fine particulate matter data is imperative, however, values are seldom complete. One of the most important issues for variance estimation of population total estimators using ratio and generalized regression estimators under unequal probability sampling without replacement is that it requires joint inclusion probability which can be difficult to find. In this paper we solve this problem by proposing new variance estimators for population total in the presence of nonresponse under unequal probability sampling without replacement under the uniform nonresponse mechanism. Two approaches are used to construct the new variance estimators; estimating the joint inclusion probability and free joint inclusion probability. Simulation studies and an application to fine particulate matter in the north of Thailand are considered in the study. The results show that the variance estimators from the latter method give a smaller relative root mean square error and relative bias than the variance estimator from the former one. Nevertheless, the proposed variance estimator with the free joint inclusion probability provides a narrower confidence interval compared to others.
References
Bacanli S, Kadilar C. Ratio estimators with unequal probability designs. Pakistan J Stat. 2008; 24(3): 167-172.
Berger YG. Rate of convergence to normal distribution for the Horvitz-Thompson estimator. J. Stat. Plan. 1998; 67 (2): 209-226.
Berger YG. A simple variance estimator for unequal probability sampling without replacement. J Appl Stat. 2003; 31: 305-315.
Berger YG, Tirari MEH, Tille Y. Towards optimal regression estimation in sample surveys. Aust N Z J Stat. 2003; 45(3): 319-329. https://doi:10.1111/1467-842X.00286.
Brewer KRW. Combined survey sampling inference: weighing Basu’s elephants, London: Arnold; 2002.
Brewer KRW, Donadio ME. The high entropy variance of the Horvitz-Thompson estimator. Surv Methodol. 2003; 29 (2): 189-196.
Estevao VM, and Särndal CE. Survey estimates by calibration on complex auxiliary information. Int Stat Rev. 2006; 74 (2): 127-147.
Hajek J. Asymptotic theory of rejective sampling with varying probabilities from a finite population. Ann Math Stat. 1964. 35 (4):1491-1523.
Hajek J. Sampling from finite population. New York: Marcel Dekker; 1981.
Hansen MH, Hurwitz WN. On the theory of sampling from finite populations. J Am Stat Assoc. 1943;14(4): 333-362.
Hartley HO, Rao JNK. Sampling with unequal probability and without replacement. J Am Stat Assoc. 1962; 33(2): 350-374.
Horvitz DF, Thompson DJ. A generalization of sampling without replacement from a finite universe. J Am Stat Assoc. 1952; 47(260): 663-685.
IQAir World Air Quality Report. Air quality and pollution city ranking. 2021 [cited 2023 Sep 6]. Available from: https://www.iqair.com/th-en/world-air-quality-ranking.
Lawson N. Variance estimation in the presence of nonresponse with free joint Inclusion probability under unequal probability sampling without replacement. Proceedings of the 3rd International Conference on Big Data and Education; 2020 Apr 1; pp. 50-54.
Lawson N, Ponkaew C. New generalized regression estimator in the presence of nonresponse under unequal probability sampling. Commun Stat Theory Methods. 2019; 48(10): 2483-2498.
Midzuno H. On the sampling system with probability proportional to sum of sizes. Ann Inst Stat Math. 1952; 553: 99-107.
Montanari GE. Post sampling efficient qr-prediction in large sample survey. Int Stat Rev. 1987; 55(2): 191-202.
Nangsue N, Berger YG. Optimal regression estimator for stratified two-stage sampling. In: Fulvia M, Luigi CP, Giovanna RM, editors. Contributions to Sampling Statistics. Cham: Springer; 2014; p. 167-177.
Pollution Control Department. Thailand’s air quality and situation reports. Bangkok, Thailand. 2022. [cited 2023 Sep 6]. Available from: http://air4thai.pcd.go.th/webV2/history/.
Ponkaew C, Lawson N. A new ratio estimator for population total in the presence of nonresponse under unequal probability sampling without replacement. Thai J. Math.: Special Issue (ACFPTO2018) on: Advances in Fixed Point Theory towards Real World Optimization Problem. 2018; 417-429.
Ponkaew C, Lawson N. Estimating variance in the presence of nonresponse under unequal probability sampling. Suranaree J Sci Technol. 2019; 26 (3): 293-302.
R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. 2021 [cited 2023 Sep 6]. Available from: https://www.R-project.org.
Särndal CE, Lundström S. Estimation in surveys with nonresponse. New York: John Wiley & Sons; 2005.
Sen AR. On the estimate of the variance in sampling with varying probabilities, J Indian Soc Agri Stat. 1953; 5: 119-127.
Sichera R. Approximate inclusion probabilities for survey sampling. R package version 0.1.2. 2020. 2020 [cited 2023 Sep 6]. Available from: URL https://CRAN.project.org/package=jipApprox.
Yates F, Grundy PM. Selection without replacement from within strata with probability proportional to size. J R Stat Soc Series B. 1953; 15: 235-261.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
