In statistics, the HorvitzâThompson estimator, named after Daniel G. Horvitz and Donovan J. Thompson, [1] is a method for estimating the total [2] and mean of a pseudo-population in a stratified sample by applying inverse probability weighting to account for the difference in the sampling distribution between the collected data and the a target population. The HorvitzâThompson estimator is frequently applied in survey analyses and can be used to account for missing data, as well as many sources of unequal selection probabilities.
Formally, let be an independent sample from n of N â„ n distinct strata with a common mean ÎŒ. Suppose further that is the inclusion probability that a randomly sampled individual in a superpopulation belongs to the ith stratum. The HorvitzâThompson estimator of the total is given by: [3]: 51
and the HorvitzâThompson estimate of the mean is given by:
In a Bayesian probabilistic framework is considered the proportion of individuals in a target population belonging to the ith stratum. Hence, could be thought of as an estimate of the complete sample of persons within the ith stratum. The HorvitzâThompson estimator can also be expressed as the limit of a weighted bootstrap resampling estimate of the mean. It can also be viewed as a special case of multiple imputation approaches. [4]
For post-stratified study designs, estimation of and are done in distinct steps. In such cases, computating the variance of is not straightforward. Resampling techniques such as the bootstrap or the jackknife can be applied to gain consistent estimates of the variance of the HorvitzâThompson estimator. [5] The "survey" package for R conducts analyses for post-stratified data using the HorvitzâThompson estimator. [6]
The HorvitzâThompson estimator can be shown to be unbiased when evaluating the expectation of the HorvitzâThompson estimator, , as follows:
The HansenâHurwitz (1943) is known to be inferior to the HorvitzâThompson (1952) strategy, associated with a number of Inclusion Probabilities Proportional to Size (IPPS) sampling procedures. [7]
In statistics, the HorvitzâThompson estimator, named after Daniel G. Horvitz and Donovan J. Thompson, [1] is a method for estimating the total [2] and mean of a pseudo-population in a stratified sample by applying inverse probability weighting to account for the difference in the sampling distribution between the collected data and the a target population. The HorvitzâThompson estimator is frequently applied in survey analyses and can be used to account for missing data, as well as many sources of unequal selection probabilities.
Formally, let be an independent sample from n of N â„ n distinct strata with a common mean ÎŒ. Suppose further that is the inclusion probability that a randomly sampled individual in a superpopulation belongs to the ith stratum. The HorvitzâThompson estimator of the total is given by: [3]: 51
and the HorvitzâThompson estimate of the mean is given by:
In a Bayesian probabilistic framework is considered the proportion of individuals in a target population belonging to the ith stratum. Hence, could be thought of as an estimate of the complete sample of persons within the ith stratum. The HorvitzâThompson estimator can also be expressed as the limit of a weighted bootstrap resampling estimate of the mean. It can also be viewed as a special case of multiple imputation approaches. [4]
For post-stratified study designs, estimation of and are done in distinct steps. In such cases, computating the variance of is not straightforward. Resampling techniques such as the bootstrap or the jackknife can be applied to gain consistent estimates of the variance of the HorvitzâThompson estimator. [5] The "survey" package for R conducts analyses for post-stratified data using the HorvitzâThompson estimator. [6]
The HorvitzâThompson estimator can be shown to be unbiased when evaluating the expectation of the HorvitzâThompson estimator, , as follows:
The HansenâHurwitz (1943) is known to be inferior to the HorvitzâThompson (1952) strategy, associated with a number of Inclusion Probabilities Proportional to Size (IPPS) sampling procedures. [7]