Thursday, March 3
|13h30||Welcome and Coffee|
SAS – Nele Coghe, Pre-Sales Consultant and Elisabeth Versailles, Academic Relationships Manager
SAS once stood for "statistical analysis system." It began at North Carolina State University as a project to analyze agricultural research. Today SAS is helping customers in all sorts of industries – from pharmaceutical companies and banks to retail and governmental entities – through innovative analytics, business intelligence and data management software and services.
In today’s changing digital landscape, the SAS Academic Program aims providing support via free software access, free certification and courses for the data scientists of tomorrow. The Program focuses further on specialized support in research and on building partnerships, hence creating bridges between academics and the industry.
UA – Prof. Dr. Dimitri Mortelmans
SAS is a powerful instrument for researchers and analysts all over the world. Many communities discuss complex and advanced topics often forgetting the SAS starters. This presentation will give you insights in the start of a SAS career: the social science student following a statistics course. Using the introduction book “SAS in research”, this presentation shows how Antwerp Social Science students are introduced in the basics of SAS before they advance to more complex structural equation and multilevel models.
Nowadays, Law Enforcement and Intelligence agencies, such as the Federal Police, face a huge amount of unstructured data they cannot deal with. At the same time, new technologies enable to handle this Big Data evolution. This presentation shows how such agencies can use Data Mining, Text Analytics and Social Network Analysis to acquire insight from Big Data to make our lives safer. All these applications can easily be mapped to a wide range of research topics.
4. Teaching Business Analytics and Big Data at a Business School: Going Beyond the Analytical Tools and Techniques
SAS – Véronique Van Vlasselaer, Analytical Consultant
While SAS offers a profound and thoroughly tested statistical environment, OpenSource software bridges the gap to new, state-of-the-art advances in statistics and data analysis, often needed by researchers. In this presentation, we will show how the researcher can exploit a variety of built-in interfaces in SAS which enables to (1) easily call and execute programs originated from OpenSource software, (2) retrieve and merge the generated output from multiple sources, and (3) further analyze and visualize the unified results.
|15h50 - 16h20||Break|
KU Leuven – Prof. dr. N. Vandaele en Ir. Catherine Decouttere
During the past ten years our research can be described as building models for various decision support problems. Methodologically, we build mathematical models of the linear and non-linear type, both deterministic and stochastic. In order to build models for research purposes, a mix of different mathematical programming environments are available. However, when it comes to industrial prototyping with the eye on upscale deployment, SAS has been proven to possess some features which makes it highly suitable: smart and robust calculation engines, strong visualization capabilities and scalable up to the big data level.
7. SAS system for the evaluation of surrogacy in clinical trials (Theophile Bigirumurame and Ziv Shkedy).
UHasselt – PhD. Theophile Bigirumurame
In clinical trials, the determination of the true endpoint or the effect of a new therapy on the true endpoint may be difficult, requiring an expensive, invasive or uncomfortable procedure. In some trials, however, the main endpoint of interest (true endpoint), for example death, is rare and/or takes a long period of time to reach. In such trials, there would be benefit in finding a more proximate endpoint ( surrogate endpoint) to determine more quickly the effect of an intervention (Burzykowski et al., 2005).
We present the new SAS system for the evaluation of surrogate endpoints in randomized clinical trials using patients data. The SAS system for surrogacy consists of a set of friendly user macros which allow the evaluation of different types of endpoints (i.e., continuous categorical, binary survival endpoints) and produce a unified and interoperable output. We demonstrate the usage and capacities of the SAS system for surrogacy using several clinical trial datasets.
Hear about the main changes that have come to SAS, as well as some of the trends you can expect in upcoming releases. We will present how SAS responds to the Internet of Things, Streaming Analytics, agile data integration and more.
KU Leuven/UA – Prof. Dr. Peter Goos
Generally, students are not enthusiastic about statistics. By making use of modern, user-friendly software such as JMP, it is, however, possible to change this, from the very first lecture on descriptive statistics onwards. In this presentation, Peter Goos will demonstrate his approach to teaching introductory statistics to business students and bio-science engineering students. First, using several examples from his lectures on descriptive statistics, he will demonstrate how easy it is to create wonderful static and dynamic graphs that provide insight into complex data sets, in the blink of an eye. Next, he will show how to use dynamic visualizations of regression models. These are called prediction profilers and help students understand, explore and exploit interaction effects and other higher order effects. By means of a data set on hearing aid parts, he will demonstrate how to make the output of a regression analysis understandable, and useful. Finally, using data from a vitamin stability experiment, he will demonstrate that the dynamic linking feature of software packages contributes to an effective communication of the results of regression analyses.
KU Leuven – Jan Ooghe, ICTS, Head Facilities for Research
The Flemish Supercomputer Center (VSC) provides High Performance Computing power to researchers. VSC is embedded in FWO and available for all researchers in Flanders.
The presentation will show you which infrastructure is available and how you can use it. And yes, SAS is available on the cluster. HPC is not only suited for large distribute fluid dynamics modelling. Large parameters sweeps or independent large scale simulations can also run on the cluster. Breaking out from the desktop to central computer resources can bring the science to a new level.