Menu
Log in


SSA SA November Branch meeting - CANCELLED

  • 27 Nov 2024
  • 6:00 PM - 7:00 PM
  • University of South Australia City East, North Tce, Centenary Building, Level 3, Room C3-22

The meeting advertised below has had to be cancelled due to unavailability of the speaker.  The talk will be rescheduled for 2025.

Apologies for the late notice.

---------------------------------------

The South Australian Branch of the Statistical Society would like to invite you to a hybrid November meeting of the 2024 program. This is both an in-person and online presentation regarding practical lessons in the use of regression by Lauren Kennedy from the University of Adelaide.

Date: 27th November (Wednesday) 2024

Time: 6:00 – 7:00 pm (Adelaide time) (Refreshments from 5:30 pm, Zoom access from 5:45pm)

Venue: University of South Australia City East, North Tce, Centenary Building, Level 3, Room C3-22.

https://unisa.edu.au/siteassets/campus-facilities/maps/city-east-campus-map.pdf

Dinner: For those who are interested, a post-meeting dinner is planned for 7:30pm at Café Michael 2, 204 Rundle St.  Please RSVP to Lan.Kelly@unisa.edu.au by 5pm Monday November 25


Topic: Things I've learnt the hard way about regression

Regression. It's one of our fundamental tools of the trade. Yet the world of examples we teach in class and the realities of real-world data are far apart. In this talk I share a collection of examples that are odd, interesting and sometimes counterintuitive. Based on my experiences working in prediction and post-stratification based survey inference, my examples include individual versus group prediction, logistic vs linear regression and rare events bias. During the discussion time I hope to leave space for you to share your own weird and wonderful examples. 


Speaker biography

Lauren is a Senior Lecturer in Statistics at the University of Adelaide. Her work focusses on survey statistics and sampling, and applied Bayesian statistics. She is particularly interested in deepening methodological understanding through applied work and consideration of the statistical workflow. 

Powered by Wild Apricot Membership Software