Follow through content

up logo ssa logo

Workshops

WORKSHOP TITLE

SHORT DESCRIPTION

Saturday 28 November 2015

Connecting the dots of data science: academia to industry

Dr Robert N. Rodriguez SAS Institute, Senior Director, Research and Development, SAS

and Dr F Kanfer, UP

The summit addresses the latest trends in the field including developments in business analytics, data driven solutions, big data and automated data sources, high performance computing and modelling unstructured data.

International experts will discuss and share experiences and novel ideas.

Key industry partners as well as academia are invited to discuss and debate the role of Data Science in industry and how academia (from diverse disciplines, not only statistics) can contribute to develop the necessary skills.

SLIDES: Hall Rodriguez Jones SAS Links  Program and Abstracts

Sunday 29 November 2015

Education Workshop: Online teaching of Statistics – special project.

Prof Chris Wild, University of Auckland, NZ

and

Prof D North, KZN University

The conference traditionally has an Education Workshop. This year it is called Statistics in Education – developing a first year online course for countrywide use. This is the start of a task team to develop a course for use in South Africa to assist all universities with the huge capacity experienced at first year level. The workshop will continue through the conference as a special session with the international expert bringing in novel ideas, and have a follow up discussion workshop on 3 December 2015 (see below). This workshop will be hands on and in a computer lab.

Text Analytics Short Course

Prof Edward Jones, Texas A&M University

Half-day: 9:00 - 12:00

Text analytics started out with simple word count analyses. At present text analytics examines contextual information. Sentiment and opinion analysis, for example, makes it possible to efficiently incorporate the opinions of thousands of customers, rather than just a few. Social media applications are, amongst other, large sources for unstructured text data. The workshop explores common methodology used to analyse large complex text data sets. Although examples are illustrated using SAS Text Miner, the general approach is software independent.

Thursday 3 December 2015

Biostatistics: Applied Meta-analysis using R

The invited plenary Prof Chen will present a workshop on meta-analysis, a very important field of statistics allowing for combining results from various statistical studies, circumventing the need for new data collection. This workshop will be hands on and in a computer lab.

Abstract: This workshop is based on the book: "Applied Meta-Analysis Using R (2013)" published by Chapman and Hall/CRC. This workshop provides a most up-to-date development and a thorough presentation of meta-analysis models for clinical trial and biomedical applications with detailed step-by-step illustrations and implementation using R.  The examples are compiled from real medical and clinical trial literatures and the analyses are illustrated by a step-by-step fashion using the most appropriate R packages and functions which should enable attendees to follow the logic and gain an understanding of the meta-analysis methods and R implementation so that they may use R to analyze their own data. 

 

Outline

Session 1:

·        Brief introduction to R

·        Overview to meta-analysis for both fixed-effects and random-effects models in meta-analysis. Real datasets in clinical trials are introduced along with two commonly used R packages of  "meta" and "rmeta"

·        Meta-analysis models for binary data, such as for risk-ratio, risk difference and odds-ratio

·        Meta-analysis models for continuous data, such as for mean difference and standardized mean difference

 

Session 2:

·        Methods to quantify heterogeneity and test the significance of heterogeneity among studies in a meta-analysis and then introduce meta-regression with R package of "metafor".

·        Meta-analysis methods for individual-patient data(IPD) analysis and meta-analysis (MA)

·        Meta-analysis methods for rare-events which is timely for clinical trials of adverse-events.

·        Multivariate meta-analysis and other relevant topics in meta-analysis.

 

Complex sampling

This workshop will be presented by Prof Steve Heeringa from the Population Studies Center, University of Michigan. Workshop Details

Merging game theory and risk analysis in optimal defense of complex stochastic systems

This workshop will be presented by Dr Gregory Levitin as an extension of the special session on the broader topic of stochastic processes. This workshop will be a satellite workshop allowing non-delegates off campus to also partake via video conferencing in order to allow for a wider audience. 

Mentorship Workshop

As part of the NRF Vulnerable Discipline grant for Statistics to SASA, all bursary holders under this grant for 2015 will be sponsored to attend this workshop which aims at developing them further as academics. There will be presentations from several experienced national and international statisticians to the students as well as break away session in smaller groups with closer contact to these mentees.

Thursday 3 December 2015 and Friday 4 December 2015

Business Statistics using SAS Enterprise guide

ANOVA, Regression, and Logistic Regression

This course is designed for SAS Enterprise Guide users who want to perform statistical analyses. The course is written for SAS Enterprise Guide 7.1 along with SAS 9.4, but students with previous SAS Enterprise Guide versions will also get value from this course. An e-course is also available for SAS Enterprise Guide 5.1 and SAS Enterprise Guide 4.3.

Learn how to:

·         generate descriptive statistics and explore data with graphs

·         perform analysis of variance

·         perform linear regression and assess the assumptions

·         use diagnostic statistics to identify potential outliers in multiple regression

·         use chi-square statistics to detect associations among categorical variables

·         fit a multiple logistic regression model.