Statistical Society of Australia Canberra Branch Meeting – Annual General Meeting
Date: Tuesday 28 March 2017
Location: College of Business and Economics [Building 26C] Allan Barton Forum, Level 2, Room 248, ANU
5.15pm Refreshments in Allan Barton Forum.
5.30pm Annual General Meeting (and continuation of refreshments) in Allan Barton Forum.
6.00pm Presentation by Robert Clark — please see details below
7.15pm After the talk, there will be a dinner at Mama’s Trattoria http://www.
Details of Annual General Meeting:
The quorum for the meeting is 15 members. Please make the effort to attend.
The following documents are linked below:
1. Agenda for the 2017 AGM
4. Budget 2017
The number of copies of documents available at the meeting will be limited. Could you please print them or load them on a hand-held device and bring them with you to the meeting. Thanks.
Details of talk:
Speaker: Associate Professor Robert Clark
Title: Obtaning relative risks from binary regression models
Epidemiologists and biostatisticians sometimes use a log link in binary regression. This allows the exponentiated regression coefficient for each covariate to be interpreted as a relative risk conditional on the remaining covariates. Relative risks are simpler to interpret than the odds ratios which arise in logistic regression. There are practical and conceptual challenges in log-link binary regression, mainly due to the requirement that probabilities be less than or equal to 1. I will talk about the history of this tension between theoretical coherence and interpretability, and illustrate with some example datasets. I will also discuss a new compromise link, which equals the log up to a cutoff and then smoothly morphs into a logit. This is joint research with Margo Barr of the Sax Institute.
Robert Clark is an Associate Professor of Statistics at the University of Wollongong. He is director of the Centre for Sample Survey Methodology which is located in the National Institute for Applied Statistics Research Australia (NIASRA). Robert spent 10 years as a director and in other positions at the Australian Bureau of Statistics, and has been an academic at the University of Wollongong for 12 years. He has an active research programme in statistical ecology and the design and analysis of complex samples. He is also a statistical consultant to a range of clients and partners in Australia and New Zealand.