enEnglish
CROS

Use of R in Official Statistics - uRos2020

Country: 
Austria
Location: 
Vienna
Date: 
Wednesday, 2 December, 2020 to Friday, 4 December, 2020

https://www.urosconf.org

https://twitter.com/@urosconf

The annual conference on the Use of R in Official Statistics will take place at the premises of Statistics Austria, 6-8 May 2020 2-4 December 2020. The conference consists of one tutorial day (6 May 2020 2 December 2020) and two days of scientific conference (7-8 May 2020 3-4 December 2020).

We are also excited to announce an “unconference” “Use of R in Official Statistics” back-to-back with the uRos2020 conference, scheduled on 4-5 May 2020 30 November - 1 December 2020.

Over the last two decades R has become the lingua franca for statisticians, methodologists and data scientists worldwide. The reasons why the official statistics community is rapidly embracing R are clear: it has an active worldwide community of users, there is wide support from the industry and it combines a vast amount of functionalities for data preparation, methodology, visualisation and application building. Moreover, R-based software is exchanged through strictly enforced technical standards: “R is probably the most thoroughly validated statistics software on Earth.” – Uwe Ligges, CRAN maintainer (useR!2017).

Finally, R is a free (libre) Open Source software product, making it an ideal environment for sharing, collaboration and industrialising official statistics.

The annual event “use of R in Official Statistics” started as a local conference at the Statistical Office of Romania in 2013, with the first international edition being held at that same office in 2014. In 2018, Statistics Netherlands hosted the sixth edition of this event, which brought together official statisticians, scientists, and prominent members of the R community to share ideas on using and developing R tools in the area of official statistics.

Now it is Statistics Austria’s turn to host this stimulating meeting and we very much look forward to welcoming you to Vienna in 2020.

 

Important dates

  • 4-5 May 2020 30 November - 1 December 2020: unconfUROS & R Crash Course
  • 6-8 May 2020 2-4 December 2020: Scientific conference & Tutorial sessions
  • 15 June 2020 15 January 2021: Journal paper submission deadline

Keynote Speakers

We are happy to announce that we have already fixed two keynote talks and one more will be announced soon.

Isabel Molina Peralta

is associate professor at the Department of Statistics of Universidad Carlos III de Madrid since 2009. She got her Ph. D. in Statistics and Operations Research in 2003 at Universidad Miguel Hernández de Elche. She has published more than 30 papers in peer-reviewed statistical journals, most of them on small area estimation, and has received several awards. She is currently associate editor for the scientific journals “Survey Methodology” and “Journal of Survey Statistics and Methodology”. She is co-author, together with J.N.K. Rao, of the Wiley book “Small area estimation, 2nd Ed.”

“Small area estimation using R, with application to poverty mapping.”

Small area estimation is a promising, but also well established and well developed family of methods to gain insights on detailed domains or areas. Not widely used in statistical offices yet, available resources in R provide an opportunity to change that. Existing R packages designed for small area estimation with a special focus on the applicability for poverty mapping are reviewed. Applications to poverty mapping are presented to illustrate some of the available functions.

Matthias Templ

made his PhD in technical mathematics and the venia legendi (habilitation) in statistics at the Vienna University of Technology (TU Wien). He was associated professor at the TU Wien and consultant at the Palacky University Olomouc. He is employed by the Institute for Data Analysis and Process Design at the Zurich University of Applied Sciences, where he teaches and research in the area of statistical modelling and data science.
His main research interests include computational statistics, compositional data analysis, imputation and statistical disclosure control. He published more than 50 papers in well-known indexed scientific journals and he is author of several R packages.
Matthias Templ is editor-in-chief of the Austrian Journal of Statistics.

“Functional data analysis in Bayes Spaces with an Application to spatio-temporal population data”

Probability density functions are frequently used to characterize the distributional properties of large-scale database systems. As functional compositions, densities carry primarily relative information. As such, standard methods of functional data analysis (FDA) are not appropriate for their statistical processing and thus a compositional alternative is proposed. The aim of this presentation is to outline a concise methodology for functional principal component analysis of densities based on the geometry of the Bayes space B2 of functional compositions. Advances of the proposed approach are demonstrated using a real-world example of population pyramids in Upper Austria. For compositional analysis we also introduce the R package robCompositions.

Kurt Hornik

is the head of the Institute for Statistics and Mathematics and the Research Institute for Computational Methods at WU Vienna. His main research interests include statistical computing, statistical and machine learning and quantitative risk management. He is a member of the R Development Core Team and hosting the central CRAN server at WU.

“Analyzing Texts with R”

Working with textual data (“text mining”) is becoming increasingly important in modern statistics. We show how R can be used for basic and advanced natural language processing tasks, manipulating collections of text documents, and deriving features/representations suitable for further statistical analyses. We illustrate how these conceptual and computational tools can be employed in several applications, including document clustering, stylometry, cultoromics, semantic lexicon induction, extracting financial sentiment, and modeling readability.