Logo Utrecht University

Circular Data Modelling

Activities

Upcoming events

May 3 2018 – M&S power snack

Circular data in social and behavioral sciences

 

 

M&S organizes a series of meetings for teachers and researchers of the Faculty of Social and Behavioural Sciences on choosing a research design, gathering data, and analyzing your data. Each lecture will start with the basics, such that specialized prior knowledge is not necessary, but will also cover recent developments and state-of-the-art applications. The power snacks are aimed at teachers and researchers, but students are welcome too!

The meeting on the third of May will be about circular data modelling, and is given by Irene Klugkist, Jolien Cremers and Kees Mulder. More information on the contents of this meeting can be found on the Utrecht University agenda.

11:30-12:30, Boothzaal, University Library Uithof. (Registration is not required.)

 

24-29 June 2018 – ISBA World Meeting Session

Bayesian methods for directional data

Directional statistics is a branch of statistics that deals with directional data that can be encountered in many scientific disciplines. The two most common forms are data that lie on the circumference of the unit circle (circular data) or on the surface of the unit hypersphere (spherical data).  Some examples are wind directions, circadian rhythms and navigational data.

A classical monograph on directional statistics appeared in 1972, written by Mardia, who is still active in the field (e.g., see Mardia and Jupp, 2000). To date, the development of directional statistics is far from finished. In the past 15-20 years, the attention for directional statistics has dramatically increased and currently there is an active community of researchers rapidly producing new methods, papers, books and software (e.g., Jammalamadaka & SenGupta, 2001; Pewsey, Neuhauser, Ruxton, 2013; Ley & Verdebout, 2017). However, the majority of this work is non-Bayesian. The proposed speakers in this session believe that the area of directional statistics will benefit from the Bayesian approach and will present some of their recent work on this topic.

Speakers

  • Jolien Cremers
  • Eduardo Gutiérrez-Peña
  • Antonello Maruotti
  • Kees Mulder

More information on the 2018 ISBA World Meeting can be found on the ISBA website.

 

9-13 July 2018 – IMPS 2018 conference presentation

One direction? On the modelling of circular data in psychology.

There are many examples of circular data in psychology, such as personality measurement, visual perception of space, cognitive maps and more. However, knowledge on this type of data is not well-spread and are often analyzed using simple one-way ANOVA designs with a circular outcome.

More complex models for circular data do exist, such as projected normal regression and mixed-effects models. We developed an R-package offering researchers new tools for analyzing and interpreting these types of models. Using example data from cognitive psychology, tools from this package and their usefulness for applied researchers will be demonstrated.

More information about this presentation, as well as the full program, can be found on the IMPS website.

 

July 28 – August 2, 2018 – Joint Statistical Meetings 2018 conference presentation

One direction? On the modelling of circular data using projected normal distributions.

At the JSM 2018, Jolien Cremers will hold a presentation (oral + poster) on methods to analyze regression and mixed-effects models with a circular outcome. This is done using projected normal distributions, a technique that allows for a smooth translation of existing linear models to the circular case. New methods to help interpret these types of models are presented, as well as an R-package (bpnreg) developed to allow applied researchers to fit projected normal regression and mixed-effect models.

More information on JSM2018  and the presentation can be found on their website.

 

May 2 2018 – Research meeting Groningen presentation

Circular methods for circumplex data

On the second of May, Jolien Cremers and Irene Klugkist will hold a presentation on the use of circular methods to analyze circumplex data. The presentation will be held at the University of Groningen.

Past events

Workshop at Utrecht University

Our first workshop on Circular Data Modelling was at Utrecht University (CDM@UU), the Netherlands on November 24-25, 2016.

The full program for this workshop can be found here.

Conference presentations

2018

  • Kees Mulder: Symposium “Connecting Mathematical methods across UU” (2018, oral).

2017

  • Irene Klugkist: ADISTA2017 in Rome (invited speaker)
  • Jolien Cremers: ADISTA2017 (2017, poster)
  • Kees Mulder: ADISTA2017 (2017, poster), CMStatistics (2017, oral), IOPS (Winter 2017, oral)

2016

  • Irene Klugkist: CDM@UU (organizer)
  • Jolien Cremers: ISBA (2016, poster), CDM@UU (2016, oral), CMStatistics (2016, oral)
  • Kees Mulder: ISBA (2016, poster), CDM@UU (2016, oral), CMStatistics (2016, oral)

2015

  • Jolien Cremers: ML congres (2015, oral), OBayes congres (2015, poster)
  • Kees Mulder: OBayes congres (2015, poster)

2014

  • Jolien Cremers: BAYSM, Young Bayesians conference (2014, poster)
  • Kees Mulder: BAYSM, Young Bayesians conference (2014, oral)

2011

  • Irene Klugkist: Evaluating order-constrained hypotheses for circular data using permutation tests (oral presentation)
    At the International Meeting of the Psychomteric Society (IMPS), Hong Kong, China