Article from NY Times
Why Do Americans Stink at Math?
When Akihiko Takahashi was a junior in college in 1978, he was like most of the other students at his university in suburban Tokyo. He had a vague sense of wanting to accomplish something but no clue what that something should be. But that spring he met a man who would become his mentor, and this relationship set the course of his entire career.
Takeshi Matsuyama was an elementary-school teacher, but like a small number of instructors in Japan, he taught not just young children but also college students who wanted to become teachers. At the university-affiliated elementary school where Matsuyama taught, he turned his classroom into a kind of laboratory, concocting and trying out new teaching ideas. When Takahashi met him, Matsuyama was in the middle of his boldest experiment yet — revolutionizing the way students learned math by radically changing the way teachers taught it.
I am pleased to announce that the Department of Statistics is now an affiliate of the Computational Science and Engineering (CSE) program.
We offer CSE certificates in Statistics for both undergraduates and PhD students. Please see the relevant links from our degree programs page:
Both the undergraduate and graduate certification programs offer scholarship/fellowship opportunities on a competitive basis for enrolled students. Please see the relevant deadlines for applications.
Douglas G. Simpson
Professor and Chair
Department of Statistics
University of Illinois at Urbana-Champaign
Which do you prefer to use, SAS or R? On her blog, Linda Burtch of BurtchWorks, an executive recruiter for those in data analysis, recently posted the results of a survey she conducted.
Check it out here: http://lindaburtch.blogspot.com/2014/04/the-deep-dive-sas-vs-r.html
Check out this Wall Street Journal article, where the accelerating need for professionals with quantitative skills is discussed, and how, “In 15 years, if you don’t have a solid quant background, you might have a permanent pink slip.”
NCSA is sponsoring a presentation by Edward Tufte titled “The Thinking Eye” on April 10, 2014 7:00 pm at Foellinger Auditorium, UIUC.
Free and open to the public.
Edward Tufte, data theorist and visualization pioneer, will discuss seeing, reasoning, producing in science and art. Topics include evidence and inference, strategies for identifying excellence, and practical advice for seeing better in the real world and on the glowing flat rectangle of the computer screen.
Edward Tufte is a statistician and landscape sculptor, and Professor Emeritus of Political Science, Statistics, and Computer Science at Yale University. He wrote and designed 4 classic books on data visualization: Beautiful Evidence, The Visual Display of Quantitative Information, Envisioning Information, and Visual Explanations. The New York Times described ET as the “Leonardo da Vinci of Data.” and Business Week as the “Galileo of Graphics.” He is currently constructing a 234-acre sculpture park in northwest Connecticut, Hogpen Hill Farms, which will show his artworks and remain open space in perpetuity. ET’s art exhibit, “The Cognitive Art of Feynman Diagrams,” will be at Fermilab from April 15 to June 26.
Has it been a while since you’ve thought about what a p-value says, and what it doesn’t say?
An article in this month’s issue of Nature features Stephen Ziliak, Economist at Roosevelt University in Chicago. The article discusses the ways that p-values are commonly misused. Ziliak is quoted as saying “P values are not doing their job, because they can’t.”
If you find this article intriguing, I recommend Ziliak’s book, “The Cult of Statistical Significance“. It’s available on Amazon and through the university library system.
Data Boom: The field of statistics is soaring—according to latest stats.
When Douglas Simpson arrived on the University of Illinois campus in 1985 to take his first professorship straight out of grad school, the time-traveling movie Back to the Future ruled the theaters and the U of I Department of Statistics had just been founded. Today, the field of statistics is booming, and that’s something no one but a time traveler, back from the future, might have predicted in 1985, when the U of I department had only a handful of students.
Fittingly, statistics back up the stunning rise in statistics.
Even as recently as 10 to 15 years ago, the Department of Statistics had only about 30 undergrads, five to 10 master’s students, and maybe 17 to 20 PhD students, Simpson says. Since then, the number of undergrads has skyrocketed to about 200, the number of master’s students has gone up eight times to over 80, and the numbr of PhD students has nearly doubled to 35.
…Read the rest of the article at http://www.las.illinois.edu/news/2013/statisticsdept/.
The University of Illinois Statistics Department is among the Top Five in the U.S. Conferring Bachelor’s Degrees in Statistics in 2012. Statistics degrees have increased greatly in recent years, with growth driven by demand for analytical skills in the workplace, according to the American Statistical Association.
The number of students earning bachelor’s degrees in statistical science has more than doubled in the last five years, topping 1,000 for the first time in 2012, says American Statistical Association (ASA) President Marie Davidian. “Availability of jobs, the diversity of jobs, starting salaries, a low unemployment rate and similar facts help influence a student’s college major decision,” says Davidian. “Combined, these factors are helping to make statistics a more attractive career choice to today’s high-school and college students.”
Full article at https://www.amstat.org/newsroom/pressreleases/Number-of-Statistics-Majors-Continues-Climbing.pdf
Nov 14, 2013
1005 Beckman Institute
|1:00 – 1:20
||Spectral Clustering in Heterogeneous Networks Using the Stochastic Blockmodel
|1:20 – 1:40
||Approximate Algorithms for Bayesian Variable Selection
|1:40 – 2:00
||Latent Space Models for Dynamic Networks
|2:00 – 2:05
|2:05 – 2:25
||Sufficient Dimension Reduction for Longitudinal Data
|2:25 – 2:45
||Unit Root Testing with Piecewise Locally Stationary Errors
||Statistics and Sabermetrics: Advanced Analytics inside Major League Baseball
Break for refreshments and presentation of awards
|3:45 – 4:15
Norton Prize WinnerStructure Detection in Networks4:15 – 5:00Yuhong Yang,
Bohrer SpeakerModel Selection, Model Combining, and Rate of Convergence for High-Dimensional Regression