### Refine

#### Keywords

- Erhebungsverfahren (2)
- Calibration (1)
- Mietpreis (1)
- Nicht-linear Statistiken (1)
- Numerical Optimization (1)
- Official Statistics (1)
- Optimal Multivariate Allocation (1)
- Optimierung (1)
- Schätztheorie (1)
- Schätzung (1)

#### Institute

The dissertation deals with methods to improve design-based and model-assisted estimation techniques for surveys in a finite population framework. The focus is on the development of the statistical methodology as well as their implementation by means of tailor-made numerical optimization strategies. In that regard, the developed methods aim at computing statistics for several potentially conflicting variables of interest at aggregated and disaggregated levels of the population on the basis of one single survey. The work can be divided into two main research questions, which are briefly explained in the following sections.
First, an optimal multivariate allocation method is developed taking into account several stratification levels. This approach results in a multi-objective optimization problem due to the simultaneous consideration of several variables of interest. In preparation for the numerical solution, several scalarization and standardization techniques are presented, which represent the different preferences of potential users. In addition, it is shown that by solving the problem scalarized with a weighted sum for all combinations of weights, the entire Pareto frontier of the original problem can be generated. By exploiting the special structure of the problem, the scalarized problems can be efficiently solved by a semismooth Newton method. In order to apply this numerical method to other scalarization techniques as well, an alternative approach is suggested, which traces the problem back to the weighted sum case. To address regional estimation quality requirements at multiple stratification levels, the potential use of upper bounds for regional variances is integrated into the method. In addition to restrictions on regional estimates, the method enables the consideration of box-constraints for the stratum-specific sample sizes, allowing minimum and maximum stratum-specific sampling fractions to be defined.
In addition to the allocation method, a generalized calibration method is developed, which is supposed to achieve coherent and efficient estimates at different stratification levels. The developed calibration method takes into account a very large number of benchmarks at different stratification levels, which may be obtained from different sources such as registers, paradata or other surveys using different estimation techniques. In order to incorporate the heterogeneous quality and the multitude of benchmarks, a relaxation of selected benchmarks is proposed. In that regard, predefined tolerances are assigned to problematic benchmarks at low aggregation levels in order to avoid an exact fulfillment. In addition, the generalized calibration method allows the use of box-constraints for the correction weights in order to avoid an extremely high variation of the weights. Furthermore, a variance estimation by means of a rescaling bootstrap is presented.
Both developed methods are analyzed and compared with existing methods in extensive simulation studies on the basis of a realistic synthetic data set of all households in Germany. Due to the similar requirements and objectives, both methods can be successively applied to a single survey in order to combine their efficiency advantages. In addition, both methods can be solved in a time-efficient manner using very comparable optimization approaches. These are based on transformations of the optimality conditions. The dimension of the resulting system of equations is ultimately independent of the dimension of the original problem, which enables the application even for very large problem instances.

Sample surveys are a widely used and cost effective tool to gain information about a population under consideration. Nowadays, there is an increasing demand not only for information on the population level but also on the level of subpopulations. For some of these subpopulations of interest, however, very small subsample sizes might occur such that the application of traditional estimation methods is not expedient. In order to provide reliable information also for those so called small areas, small area estimation (SAE) methods combine auxiliary information and the sample data via a statistical model.
The present thesis deals, among other aspects, with the development of highly flexible and close to reality small area models. For this purpose, the penalized spline method is adequately modified which allows to determine the model parameters via the solution of an unconstrained optimization problem. Due to this optimization framework, the incorporation of shape constraints into the modeling process is achieved in terms of additional linear inequality constraints on the optimization problem. This results in small area estimators that allow for both the utilization of the penalized spline method as a highly flexible modeling technique and the incorporation of arbitrary shape constraints on the underlying P-spline function.
In order to incorporate multiple covariates, a tensor product approach is employed to extend the penalized spline method to multiple input variables. This leads to high-dimensional optimization problems for which naive solution algorithms yield an unjustifiable complexity in terms of runtime and in terms of memory requirements. By exploiting the underlying tensor nature, the present thesis provides adequate computationally efficient solution algorithms for the considered optimization problems and the related memory efficient, i.e. matrix-free, implementations. The crucial point thereby is the (repetitive) application of a matrix-free conjugated gradient method, whose runtime is drastically reduced by a matrx-free multigrid preconditioner.

Mittels Querschnittserhebungen ist es möglich Populationsparameter zu einem bestimmten Zeitpunkt zu schätzen. Jedoch ist meist die Veränderung von Populationsparametern von besonderem Interesse. So ist es zur Evaluation von politischen Zielvorgaben erforderlich die Veränderung von Indikatoren, wie Armutsmaßen, über die Zeit zu verfolgen. Um zu testen ob eine gemessene Veränderung sich signifikant von Null unterscheidet bedarf es einer Varianzschätzung für Veränderungen von Querschnitten. In diesem Zusammenhang ergeben sich oft zwei Probleme; Zum einen sind die relevanten Statistiken meist nicht-linear und zum anderen basieren die untersuchten Querschnittserhebungen auf Stichproben die nicht unabhängig voneinander gezogen wurden. Ziel der vorliegenden Dissertation ist es einen theoretischen Rahmen zur Herleitung und Schätzung der Varianz einer geschätzten Veränderung von nicht-linearen Statistiken zu geben. Hierzu werden die Eigenschaften von Stichprobendesigns erarbeitetet, die zur Koordination von Stichprobenziehungen in einer zeitlichen Abfolge verwendet werden. Insbesondere werden Ziehungsalgorithmen zur Koordination von Stichproben vorgestellt, erarbeitet und deren Eigenschaften beschrieben. Die Problematik der Varianzschätzung im Querschnitt für nicht-lineare Schätzer bei komplexen Stichprobendesigns wird ebenfalls behandelt. Schließlich wird ein allgemeiner Ansatz zur Schätzung von Veränderungen aufgezeigt und es werden Varianzschätzer für die Veränderung von Querschnittschätzern basierend auf koordinierten Querschnittstichproben untersucht. Insbesondere dem Fall einer sich über die Zeit verändernden Population wird eine besondere Bedeutung im Rahmen der Arbeit beigemessen, da diese im Anwendungsfall die Regel darstellen.

A basic assumption of standard small area models is that the statistic of interest can be modelled through a linear mixed model with common model parameters for all areas in the study. The model can then be used to stabilize estimation. In some applications, however, there may be different subgroups of areas, with specific relationships between the response variable and auxiliary information. In this case, using a distinct model for each subgroup would be more appropriate than employing one model for all observations. If no suitable natural clustering variable exists, finite mixture regression models may represent a solution that „lets the data decide“ how to partition areas into subgroups. In this framework, a set of two or more different models is specified, and the estimation of subgroup-specific model parameters is performed simultaneously to estimating subgroup identity, or the probability of subgroup identity, for each area. Finite mixture models thus offer a fexible approach to accounting for unobserved heterogeneity. Therefore, in this thesis, finite mixtures of small area models are proposed to account for the existence of latent subgroups of areas in small area estimation. More specifically, it is assumed that the statistic of interest is appropriately modelled by a mixture of K linear mixed models. Both mixtures of standard unit-level and standard area-level models are considered as special cases. The estimation of mixing proportions, area-specific probabilities of subgroup identity and the K sets of model parameters via the EM algorithm for mixtures of mixed models is described. Eventually, a finite mixture small area estimator is formulated as a weighted mean of predictions from model 1 to K, with weights given by the area-specific probabilities of subgroup identity.