Austrian Journal of Statistics AUSTRIAN STATISTICAL SOCIETY Volume 45, Number 1, 2016 Special Issue on R Österreichische Zeitschrift für Statistik ÖSTERREICHISCHE STATISTISCHE GESELLSCHAFT Austrian Journal of Statistics; Information and Instructions GENERAL NOTES The Austrian Journal of Statistics is an open-access journal with a long history and is published appro- ximately quarterly by the Austrian Statistical Society. Its general objective is to promote and extend the use of statistical methods in all kind of theoretical and applied disciplines. Special emphasis is on methods and results in official statistics. Original papers and review articles in English will be published in the Austrian Journal of Statistics if judged consistently with these general aims. All papers will be refereed. Special topics sections will appear from time to time. Each section will have as a theme a specialized area of statistical application, theory, or methodology. Technical notes or problems for considerations under Shorter Communications are also invited. A special section is reserved for book reviews. All published manuscripts are available at http://www.ajs.or.at (old editions can be found at http://www.stat.tugraz.at/AJS/Editions.html) Members of the Austrian Statistical Society receive a copy of the Journal free of charge. To apply for a membership, see the website of the Society. Articles will also be made available through the web. PEER REVIEW PROCESS All contributions will be anonymously refereed which is also for the authors in order to getting positive feedback and constructive suggestions from other qualified people. Editor and referees must trust that the contribution has not been submitted for publication at the same time at another place. It is fair that the submitting author notifies if an earlier version has already been submitted somewhere before. Manuscripts stay with the publisher and referees. The refereeing and publishing in the Austrian Journal of Statistics is free of charge. The publisher, the Austrian Statistical Society requires a grant of copyright from authors in order to effectively publish and distribute this journal worldwide. OPEN ACCESS POLICY This journal provides immediate open access to its content on the principle that making research freely available to the public supports a greater global exchange of knowledge. ONLINE SUBMISSIONS Already have a Username/Password for Austrian Journal of Statistics? Go to http://www.ajs.or.at/index.php/ajs/login Need a Username/Password? Go to http://www.ajs.or.at/index.php/ajs/user/register Registration and login are required to submit items and to check the status of current submissions. AUTHOR GUIDELINES The original LATEX-file guidelinesAJS.zip (available online) should be used as a template for the setting up of a text to be submitted in computer readable form. Other formats are only accepted rarely. SUBMISSION PREPARATION CHECKLIST The submission has not been previously published, nor is it before another journal for consideration • (or an explanation has been provided in Comments to the Editor). The submission file is preferable in LATEX file format provided by the journal. • All illustrations, figures, and tables are placed within the text at the appropriate points, rather than • at the end. The text adheres to the stylistic and bibliographic requirements outlined in the Author Guidelines, • which is found in About the Journal. COPYRIGHT NOTICE The author(s) retain any copyright on the submitted material. The contributors grant the journal the right to publish, distribute, index, archive and publicly display the article (and the abstract) in printed, electronic or any other form. Manuscripts should be unpublished and not be under consideration for publication elsewhere. By sub- mitting an article, the author(s) certify that the article is their original work, that they have the right to submit the article for publication, and that they can grant the above license. Austrian Journal of Statistics Volume 45, Number 1, 2016 Special Guest Editors: Andreas ALFONS, Rainer STÜTZ Editor-in-chief: Matthias TEMPL http://www.ajs.or.at Published by the AUSTRIAN STATISTICAL SOCIETY http://www.osg.or.at Österreichische Zeitschrift für Statistik Jahrgang 45, Heft 1, 2016 ÖSTERREICHISCHE STATISTISCHE GESELLSCHAFT Impressum Editor: Matthias Templ, Statistics Austria & Vienna University of Technology Editorial Board: Peter Filzmoser, Vienna University of Technology Herwig Friedl, TU Graz Bernd Genser, University of Konstanz Peter Hackl, Vienna University of Economics, Austria Wolfgang Huf, Medical University of Vienna, Center for Medical Physics and Biomedical Engineering Alexander Kowarik, Statistics Austria, Austria Johannes Ledolter, Institute for Statistics and Mathematics, Wirtschaftsuniversität Wien & Management Sciences, University of Iowa Werner Müller, Johannes Kepler University Linz, Austria Josef Richter, University of Innsbruck Milan Stehlik, Department of Applied Statistics, Johannes Kepler University, Linz, Austria Wolfgang Trutschnig, Department for Mathematics, University of Salzburg Regina Tüchler, Austrian Federal Economic Chamber, Austria Helga Wagner, Johannes Kepler University Linz, Austria Walter Zwirner, University of Calgary, Canada Book Reviews: Ernst Stadlober, Graz University of Technology Printed by Statistics Austria, A-1110 Vienna Published approximately quarterly by the Austrian Statistical Society, C/o Statistik Austria Guglgasse 13, A–1110 Wien c Austrian Statistical Society Further use of excerpts only allowed with citation. All rights reserved. Contents Page Andreas ALFONS, Rainer STÜTZ: Editorial . 1 Marc BILL, Beat HULLIGER: Treatment of Multivariate Outliers in Incomplete Business Survey Data . 3 Kevin JAKOB, Matthias FISCHER: GCPM: A Flexible Package to Explore Credit Portfolio Risk . 25 Jan-Philipp KOLB: Geovisualisation: Possibilities with R .................... 45 Sebastian WARNHOLZ, Timo SCHMID: Simulation Tools for Small Area Estimation: Introducing the R Package saeSim.............................. 55 Andreas ALFONS, Christophe CROUX, Peter FILZMOSER: Robust Maximum Association Between Data Sets: The R Package ccaPP ...................... 71 Thomas MENDLIK, Georg HEINRICH, Andreas GOBIET, Armin LEUPRECHT: From Climate Simulations to Statistics – Introducing the wux Package . 81 Matthias TEMPL, Valentin TODOROV: The Software Environment R for Official Statistics and Survey Methodology . 97 1 Editorial Since its early beginnings in 1993, the statistical computing environment and program- ming language R has grown rapidly and has been the lingua franca of the statistics rese- arch community for many years. Besides being embraced by the research community, R is also widely used in, e.g., IT companies, the banking and insurance sector, the biotech and pharma industry, and even newspapers and periodicals. Much of R’s success can be attributed to its open source implementation, the fact that it is freely available, as well as its possibility to be extended with user-contributed packages. This allowed an active community to grow, resulting in currently more than 9000 packages on the communi- ty package repositories CRAN (the Comprehensive R Archive Network, http://CRAN. R-project.org) and Bioconductor (http://www.bioconductor.org), and many mo- re on development platforms such as R-Forge (http://R-Forge.R-project.org) and GitHub (http://www.GitHub.com). The versatility of R is also reflected in the wide array of topics that are covered by the articles in this Special Issue. In the paper “Treatment of Multivariate Outliers in Incomplete Business Survey Data”, Marc Bill and Beat Hulliger shed some light onto the difficult task of outlier detection with complex survey data in the presence of missing values. The paper guides users of the R package modi through every step of such an analysis, demonstrating how to apply different functions for outlier detection and imputation that are available in the package. The following paper, “GCPM: A Flexible Package to Explore Credit Portfolio Risk” by Kevin Jakob and Matthias Fischer, illustrates how the package GCPM can be used to model credit portfolio risk in the banking industry. The package furthermore allows to perform sensitivity analysis with respect to model assumptions, and offers several approa- ches to speed up computations. In the paper “Geovisualisation: Possibilities with R”, Jan-Philip Kolb shows how to use R to download and modify geographical information from the community driven online map service OpenStreetMap (OSM). Moreover, the paper demonstrates how to incorpo- rate this geographical information into R’s powerful facilities for data visualization. The paper “Simulation Tools for Small Area Estimation: Introducing the R Package saeSim” by Sebastian Warnholz and Timo Schmid presents a simulation platform that is designed specifically for research in small area estimation. The key feature of the package saeSim is that every step in a simulation study is treated as a data manipulation process, which allows to write code for reproducible research that is easy to read and to modify. In the paper “Robust Maximum Association Between Data Sets: The R Package ccaPP”, Andreas Alfons, Christophe Croux and Peter Filzmoser illustrate how to use the package ccaPP to compute intuitive projection-based measures of association between data sets. Furthermore, they show how to asses the significance of those measures via permutation tests and analyze their computation time. 2 Thomas Mendlik, Georg Heinrich, Andreas