Brag Meeting 7th June 2018

When: Thursday 7th June 2018, 13:00-14:00.

Location: GP-Y801.

Included in meeting: 

 

Hello Everyone,

The fortnightly BRAG meeting will be held tomorrow, 7/6/2018, Thursday at 1.00 pm in Y801. This week we will have talks from Daniel and Jagath.

Daniel’s Talk:  QUT ACEMS book club: A summary and highlights from reading R for Data Science by Wickham & Grolemund

Abstract :   The use of R is ubiquitous in current data science, and much of its utility in this space comes from relatively new packages, which greatly reduce the difficulty in learning and interpreting R code. We have been reading the book, R for Data Science (O’Reilly Media Inc, 2016) by renowned R package author Hadley Wickham and well-known R programming educator Garrett Grolemund. The book seeks to provide a comprehensive introduction to the practice of data science using cutting edge tools from packages such as tidyverse, ggplot2 and Rmarkdown.

In this talk, I will outline the overall philosophy of the book, and give an overview of the material, emphasising the sections with content likely to be of interest to BRAG researchers. I’ll give some highlights and limitations of the book for myself and the other book club members, as well as reflections on the value and benefits of running reading groups such as ours.

Jagath’s Talk:  Multi-purpose Bayesian design for Copula models

Abstract: Design of experiments involves making decisions on how to optimally allocate available resources to gain information on a specific problem or question of interest. This is achieved by maximising a utility function through different allocations of such available resources. However, finding an optimal design for more than one objective is often not considered in the experimental design context due to lack of multi-purpose utilities that appropriately combine experimental goals. In this talk, I will discuss the available multi-purpose utilities and how these utilities can be used for design selection, particularly when the experiment involves multiple dependent responses whose joint distribution is difficult to explain using a standard multivariate distribution.

We look forward to seeing you all there!

Regards

Farzana and Trish

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s