Watch live stream of data mining expert Dick De Veaux's talk in Paris
Richard De Veaux gives a presentation on predictive analytics Sept. 28 in Paris. Sign up to watch the live stream of this event.Dick De Veaux is one of the liveliest speakers you may ever see on the topic of statistics. Here's what one person who attended a seminar he led said about him:
“I have never laughed so hard so often at a statistics presentation. He managed to keep his audience’s attention with great humor while still providing more education and information than entertainment."
On Sept. 28, Dick will give a presentation on predictive analytics in Paris at our Explorers in-person seminar. But you don't have to be in Paris to see the presentation. Before you go on summer holiday, be sure to sign up to watch the live stream online of this event.
A Professor in the Department of Mathematics and Statistics at Williams College in Williamstown, Massachusetts, Dick is an expert in applied statistics. His professional interests include data mining methodology and its application to problems in science and industry, as well as model selection and other problems for large data sets. He has been a speaker at JMP Discovery Summit and JMP Explorers seminars in the US and Europe.
Dick was kind enough to answer a few questions in advance of his Paris talk.
Can you give us a sense of what you will be talking about in your seminar?
Big data, analytics, data mining, machine learning, AI (artificial intelligence). All of these promise enormous benefits for those willing to plunge in. And they do have potential to provide insights and knowledge, when used correctly. Through a series of case studies, I’ll try to point out the major challenges and the keys to being successful.
Why is data preparation prior to analysis necessary and important?
Data are coming in at an enormous rate and in all sorts of forms, but not always of the same type and quality. In order to get the most out of any model, the data have to be optimized. Missing data is only one of the challenges of preparing data. We’ll talk about strategies for making this potentially onerous task manageable.
What types of problems or questions are best suited to predictive models?
To be suited to a predictive model, you need two things — a well-defined question and appropriate data. We’ll look at case studies to see how to ensure that you have both.
What kinds of data do you find most interesting to work with?
My personal favorite data come from masters events in running and swimming. As older athletes compete at higher and higher level, we have more and more information on the potential of aging athletes. In the last year, a new 105-110 age group has emerged for the 100 meter dash!
Looking back on your life, what would you describe as your biggest success?
My biggest success is as a communicator of statistical ideas. My series of textbooks in statistics is approaching 1 million books sold which means that we’ve had an impact on millions of young people as they embark on their analytics careers. I’m very proud of how we’ve tried to make statistics and analytics exciting and relevant to a new generation of students.