Search for: All records

Analysis of non-probability survey samples requires auxiliary information at the population level. Such information may also be obtained from an existing probability survey sample from the same finite population. Mass imputation has been used in practice for combining non-probability and probability survey samples and making inferences on the parameters of interest using the information collected only in the non-probability sample for the study variables. Under the assumption that the conditional mean function from the non-probability sample can be transported to the probability sample, we establish the consistency of the mass imputation estimator and derive its asymptotic variance formula. Variance estimators are developed using either linearization or bootstrap. Finite sample performances of the mass imputation estimator are investigated through simulation studies. We also address important practical issues of the method through the analysis of a real-world non-probability survey sample collected by the Pew Research Centre.

 

Many large‐scale surveys collect both discrete and continuous variables. Small‐area estimates may be desired for means of continuous variables, proportions in each level of a categorical variable, or for domain means defined as the mean of the continuous variable for each level of the categorical variable. In this paper, we introduce a conditionally specified bivariate mixed‐effects model for small‐area estimation, and provide a necessary and sufficient condition under which the conditional distributions render a valid joint distribution. The conditional specification allows better model interpretation. We use the valid joint distribution to calculate empirical Bayes predictors and use the parametric bootstrap to estimate the mean squared error. Simulation studies demonstrate the superior performance of the bivariate mixed‐effects model relative to univariate model estimators. We apply the bivariate mixed‐effects model to construct estimates for small watersheds using data from the Conservation Effects Assessment Project, a survey developed to quantify the environmental impacts of conservation efforts. We construct predictors of mean sediment loss, the proportion of land where the soil loss tolerance is exceeded, and the average sediment loss on land where the soil loss tolerance is exceeded. In the data analysis, the bivariate mixed‐effects model leads to more scientifically interpretable estimates of domain means than those based on two independent univariate models.

 

Data integration combining a probability sample with another nonprobability sample is an emerging area of research in survey sampling. We consider the case when the study variable of interest is measured only in the nonprobability sample, but comparable auxiliary information is available for both data sources. We consider mass imputation for the probability sample using the nonprobability data as the training set for imputation. The parametric mass imputation is sensitive to parametric model assumptions. To develop improved and robust methods, we consider nonparametric mass imputation for data integration. In particular, we consider kernel smoothing for a low-dimensional covariate and generalized additive models for a relatively high-dimensional covariate for imputation. Asymptotic theories and variance estimation are developed. Simulation studies and real applications show the benefits of our proposed methods over parametric counterparts.

 

Imputation is a popular technique for handling item nonresponse. Parametric imputation is based on a parametric model for imputation and is not robust against the failure of the imputation model. Nonparametric imputation is fully robust but is not applicable when the dimension of covariates is large due to the curse of dimensionality. Semiparametric imputation is another robust imputation based on a flexible model where the number of model parameters can increase with the sample size. In this paper, we propose a new semiparametric imputation based on a more flexible model assumption than the Gaussian mixture model. In the proposed mixture model, we assume a conditional Gaussian model for the study variable given the auxiliary variables, but the marginal distribution of the auxiliary variables is not necessarily Gaussian. The proposed mixture model is more flexible and achieves a better approximation than the Gaussian mixture models. The proposed method is applicable to high‐dimensional covariate problem by including a penalty function in the conditional log‐likelihood function. The proposed method is applied to the 2017 Korean Household Income and Expenditure Survey conducted by Statistics Korea.

 

The statistical challenges in using big data for making valid statistical inference in the finite population have been well documented in literature. These challenges are due primarily to statistical bias arising from under‐coverage in the big data source to represent the population of interest and measurement errors in the variables available in the data set. By stratifying the population into a big data stratum and a missing data stratum, we can estimate the missing data stratum by using a fully responding probability sample and hence the population as a whole by using a data integration estimator. By expressing the data integration estimator as a regression estimator, we can handle measurement errors in the variables in big data and also in the probability sample. We also propose a fully nonparametric classification method for identifying the overlapping units and develop a bias‐corrected data integration estimator under misclassification errors. Finally, we develop a two‐step regression data integration estimator to deal with measurement errors in the probability sample. An advantage of the approach advocated in this paper is that we do not have to make unrealistic missing‐at‐random assumptions for the methods to work. The proposed method is applied to the real data example using 2015–2016 Australian Agricultural Census data.

 

Many variables of interest in agricultural or economical surveys have skewed distributions and can equal zero. Our data are measures of sheet and rill erosion called Revised Universal Soil Loss Equation‐2 (RUSLE2). Small area estimates of mean RUSLE2 erosion are of interest. We use a zero‐inflated lognormal mixed effects model for small area estimation. The model combines a unit‐level lognormal model for the positive RUSLE2 responses with a unit‐level logistic mixed effects model for the binary indicator that the response is nonzero. In the Conservation Effects Assessment Project (CEAP) data, counties with a higher probability of nonzero responses also tend to have a higher mean among the positive RUSLE2 values. We capture this property of the data through an assumption that the pair of random effects for a county are correlated. We develop empirical Bayes (EB) small area predictors and a bootstrap estimator of the mean squared error (MSE). In simulations, the proposed predictor is superior to simpler alternatives. We then apply the method to construct EB predictors of mean RUSLE2 erosion for South Dakota counties. To obtain auxiliary variables for the population of cropland in South Dakota, we integrate a satellite‐derived land cover map with a geographic database of soil properties. We provide an R Shiny application calledviscover(available athttps://lyux.shinyapps.io/viscover/) to visualize the overlay operations required to construct the covariates. On the basis of bootstrap estimates of the mean square error, we conclude that the EB predictors of mean RUSLE2 erosion are superior to direct estimators.

 

A within-cluster resampling method is proposed for fitting a multilevel model in the presence of informative cluster size. Our method is based on the idea of removing the information in the cluster sizes by drawing bootstrap samples which contain a fixed number of observations from each cluster. We then estimate the parameters by maximizing an average, over the bootstrap samples, of a suitable composite loglikelihood. The consistency of the proposed estimator is shown and does not require that the correct model for cluster size is specified. We give an estimator of the covariance matrix of the proposed estimator, and a test for the noninformativeness of the cluster sizes. A simulation study shows, as in Neuhaus & McCulloch (2011), that the standard maximum likelihood estimator exhibits little bias for some regression coefficients. However, for those parameters which exhibit nonnegligible bias, the proposed method is successful in correcting for this bias.

 

Statistical and administrative agencies often collect information on related parameters. Discrepancies between estimates from distinct data sources can arise due to differences in definitions, reference periods, and data collection protocols. Integrating statistical data with administrative data is appealing for saving data collection costs, reducing respondent burden, and improving the coherence of estimates produced by statistical and administrative agencies. Model based techniques, such as small area estimation and measurement error models, for combining multiple data sources have benefits of transparency, reproducibility, and the ability to provide an estimated uncertainty. Issues associated with integrating statistical data with administrative data are discussed in the context of data from Namibia. The national statistical agency in Namibia produces estimates of crop area using data from probability samples. Simultaneously, the Namibia Ministry of Agriculture, Water, and Forestry obtains crop area estimates through extension programs. We illustrate the use of a structural measurement error model for the purpose of synthesizing the administrative and survey data to form a unified estimate of crop area. Limitations on the available data preclude us from conducting a genuine, thorough application. Nonetheless, our illustration of methodology holds potential use for a general practitioner.