Maynooth University is pleased to host the 41st Conference on Applied Statistics in Ireland on Wednesday May 12th (short course) and Thursday May 13th (main conference). This is a virtual event where details of attendance will be sent to registered participants only. There is no registration cost. See www.casi.ie for further details.
Can computers relieve data analysts of the arduous task of graphically diagnosing models?
Computer vision has come a long way in recent years. The models produced can now be used to automatically inspect the quality of items emerging along production lines, identify objects in photos and even navigate an autonomous vehicle.
Despite the fact that visualisation plays a major role in data analysis, the use and interpretation of graphics by data scientists/statisticians is subjective. Analysts rely almost entirely on their own judgement, years of experience and an implicit calculation of uncertainty when interpreting graphics. Considering data plots as a type of statistic encourages towards an inferential approach to reading data plots. By formalising data visualisation in this way, we can explore the possibility of training a computer vision model to do this visual inference task.
In this talk, I will give an introduction to these ideas and then present the results of computer vision model for evaluating residual plots, used for diagnosing statistical model fits, comparing them to human evaluations of the same plots. Who do you think wins?
The School of Mathematics and Physics at Queen’s University Belfast is pleased to host the 40th Conference on Applied Statistics in Ireland on Thursday June 25th which will be a virtual event. This is a free event, but registration is necessary. See www.casi.ie for further details.
The following topics will be covered: (1) Rationale for cluster trials; (2) Implications of clustering for sample size, and analysis; (3) RCT designs (parallel 2-arm, multi -arm, factorial, stepped wedge, balanced incomplete block designs); (4) Practical considerations when planning and conducting cluster RCTs; (5) Enhancing the informativeness of cluster RCTs; (6) Ethics of CRCTS; (7) Reporting CRCTs.
The inaugural meeting of the Young Statisticians’ Section of the Irish Statistical Association will gather together early career statisticians from across Ireland for a series of talks on career development in academia and industry. Keynote talks will be given by established professionals in both fields with the remainder of the two sessions devoted to showcasing some of the research projects conducted in universities across Ireland by postgraduate and post-doctoral researchers.
The event is open to all early career statisticians (including postgraduate and PhD students, postdoctoral researchers and lecturers) and all ISA members. There is no charge for attending this event. To register please use the Eventbrite link by Friday, 27 September.
Dr. Norma Bargary, Lecturer in Statistics, University of Limerick.
Dr. Aidan Boland, Senior Data Scientist, Edge by Ascential.
It will present an introduction and recent advances in statistical network modeling, with a special emphasis on main concepts of network analysis, sparse inference, high-dimensional data and graphical models. It will address theoretical, methodological as well as applied and practical aspects of this field.
There is no charge for attending the workshop. Priority will be given to ISA members in the case of over subscription. To register your interest, email firstname.lastname@example.org by Friday June 14, 2019. Please ensure “ISA-short course: NASM” appears in the subject line. Your place at the workshop will be confirmed via return email.
The invited speakers are:
Prof. John Hinde (National University of Ireland, Galway): Count Data Regression Models: Properties, Applications and Extensions
Prof. Sally McClean (Ulster University): Process Modelling, Process Mining and Big Data
Dr. Cóilín Minto (Galway-Mayo Institute of Technology): Challenges of missing data in fisheries science
Dr. Amirhossein Jalali (University College Dublin): Interactive tools in data and model visualisation
Dr. Kevin Hayes (University College Cork): Evaluating the utility of effect size based inference
Call for Invited Sessions – Submission Deadline Monday 26th November
The RSS Conference Board is keen to receive proposals for sessions for the RSS 2019 International Conference which cover new research and results, areas of controversy/debate, topical discussion, and informative and innovative case studies. Every year, over 600 statisticians and data scientists gather from all sectors and from over 30 countries to share information and network, attracted by a varied programme of talks and workshops. Further information on the call for invited session proposals can be found here.