No abstract

When several frequency tables need to be produced from multiple data sources, there is a risk of numerically inconsistent results. This means that different estimates are produced for the same cells or marginal totals in multiple tables. As inconsistencies of this kind are often not tolerated, there is a clear need for compilation methods for achieving numerically consistent output. Statistics Netherlands developed a Repeated Weighing (RW) method for this purpose. The scope of applicability of this method is however limited by several known estimation problems. This paper presents two new Divide-and-Conquer (D&C) methods, based on quadratic programming (QP) that avoid many of the problems experienced with RW.

The generalized multivariate Denton model for achieving consistency between large accounting frameworks developed at Statistics Netherlands (SN) is originally intended for benchmarking of supply and use (SU) tables of national accounts. The success of this application in the production process triggered using the model for other processes within the office. Currently, at SN, many production processes of national accounts, but also other departments, use the modifications of this optimization model for achieving consistency of data obtained from different sources. These include reconciliation and benchmarking of SU tables and of institutional sector accounts, ESA (European system of accounts), 1 revisions of SU tables, benchmarking of gross fixed capital formation, Population Census tables, and energy statistics figures. The mathematical model is based on a quadratic optimization function and combines different features, such as linear constraints, ratio constraints, weights, soft and hard constraints, and inequalities. The optimization problems we deal with can be very large, consisting of 500,000 variables and over 100,000 constraints. This optimization problem is solved using the commercially available software tool XPRESS and the free software tool R. For the reconciliation of trade and transport statistics, similar optimization techniques are used. In this paper, we give an overview of production processes at SN using macro-integration techniques.

Macro-integration technique is a well established method for<br />reconciliation of large, high-dimensional tables, especially applied to macroeconomic data at national statistical oces (NSO). This technique is mainly used when data obtained from dierent sources should be reconciled on a macro level. New areas of applications for this technique arise as new data sources become available to NSO's. Often these new data sources cannot be combined on a micro level, while macro integration could provide a solution for such problems. Yet, more research should be carried out to investigate if in such situations macro integration could indeed be applied. In this paper we propose two applications of macro-integration techniques in other domains than the traditional macro-economic applications. In particular: reconciliation of tables of a virtual census and reconciliation of monthly series of short term statistics gures with the quarterly gures of structural business statistics.

This article discusses methods for evaluating the variance of estimated frequency tables based on mass imputation. We consider a general set-up in which data may be available from both administrative sources and a sample survey. Mass imputation involves predicting the missing values of a target variable for the entire population. The motivating application for this article is the Dutch virtual population census, for which it has been proposed to use mass imputation to estimate tables involving educational attainment. We present a new analytical design-based variance estimator for a frequency table based on mass imputation. We also discuss a more general bootstrap method that can be used to estimate this variance. Both approaches are compared in a simulation study on artificial data and in an application to real data of the Dutch census of 2011.