February 17, 2011

Science Magazine Special Data Issue

http://www.sciencemag.org/site/special/data

Posted by kshedden at 09:29 PM | Comments (0)

Sergey Brin and data mining

An interesting (somewhat sensationalistic) article:

http://www.wired.com/magazine/2010/06/ff_sergeys_search/all/1

Posted by kshedden at 09:20 PM | Comments (0)

Google and data formats

Read about Google's new data visualization site, and the data format they are promoting:

http://www.slate.com/id/2285354

Posted by kshedden at 09:05 PM | Comments (0)

Journalism in the Age of Data

An excellent series of videos:

http://datajournalism.stanford.edu

Posted by kshedden at 09:03 PM | Comments (0)

January 12, 2011

Statistics and ESP

This article:

http://www.nytimes.com/2011/01/06/science/06esp.html?_r=1&hp

has provoked many strong responses, including this:

http://www.stat.columbia.edu/~cook/movabletype/archives/2011/01/that_silly_esp.html

Posted by kshedden at 09:43 AM | Comments (0)

December 13, 2010

An R "meta blog"

There are quite a few interesting things here if you poke around:

http://www.r-bloggers.com/

Posted by kshedden at 07:43 PM | Comments (0)

Create your own job

Think outside the box about things you might do when you graduate.

With all the new ways data are being collected and analyzed, there are many opportunities for statisticians to become entrepreneurs.

http://www.nytimes.com/2010/12/12/business/12yec.html?_r=1&src=me&ref=homepage

Posted by kshedden at 07:42 PM | Comments (0)

December 08, 2010

The Joy of Stats

An excellent series of videos about statistics:

http://www.open.ac.uk/openlearn/whats-on/the-joy-stats

Posted by kshedden at 10:40 PM | Comments (0)

Publication bias

Interesting writing about publication bias. Unfortunately you need to buy the magazine to get the full article.

http://www.newyorker.com/reporting/2010/12/13/101213fa_fact_lehrer

http://www.wired.com/wiredscience/2010/12/the-truth-wears-off/

Posted by kshedden at 10:36 PM | Comments (0)

December 04, 2010

Important and interesting applications of statistics

The second of these is likely to become an area with a lot of opportunities for statisticians.


http://www.nytimes.com/2010/12/04/books/04victorian.html?hp

http://www.nytimes.com/2010/12/04/education/04teacher.html?hp

Posted by kshedden at 10:11 AM | Comments (0)

November 29, 2010

Readings about the census

Here are a series of recent articles about the US census. There are a lot of statistical issues in the census, and also a lot of non-statistical issues. But it's all interesting.

http://www.sciencemag.org/content/304/5676/1452.full?sid=4ad2c618-f032-4669-ace7-9eb882393ec4

http://www.sciencemag.org/content/330/6002/310.full?sid=4ad2c618-f032-4669-ace7-9eb882393ec4

http://www.nature.com/nature/journal/v466/n7306/full/466532a.html

http://www.nature.com/nature/journal/v466/n7310/full/4661043a.html

And for an even greater challenge, consider the census in China:

http://www.sciencemag.org/content/330/6003/436.full?sid=0d0a2deb-292b-4fe3-b7c5-6132c483f353

or a census of the oceans!

http://www.significancemagazine.org/details/magazine/879803/The-sea-the-Census-and-statistics-.html

Posted by kshedden at 03:42 PM | Comments (0)

November 23, 2010

Good advice for writing involving numbers

http://topics.blogs.nytimes.com/2010/11/23/numbers-in-the-news/

Posted by kshedden at 11:40 AM | Comments (0)

November 17, 2010

Data for a better planet

http://www.slate.com/id/2274809/

Posted by kshedden at 09:22 AM | Comments (0)

November 05, 2010

Election poll summary

Some interesting comments about how the election polls performed this year:

http://fivethirtyeight.blogs.nytimes.com/2010/11/04/rasmussen-polls-were-biased-and-inaccurate-quinnipiac-surveyusa-performed-strongly/?hp

Posted by kshedden at 09:44 AM | Comments (0)

October 30, 2010

Article on statistics in the journal "Nature"

Nature is a leading scientific journal. Here is a recent commentary on the increasingly important role played by statisticians in scientific research:


http://www.nature.com/nature/journal/v467/n7318/full/467914a.html

Posted by kshedden at 02:13 PM | Comments (0)

October 24, 2010

Probability and statistics in the NYT

http://opinionator.blogs.nytimes.com/2010/10/24/stories-vs-statistics/?hp

Posted by kshedden at 09:28 PM | Comments (0)

October 21, 2010

Article about election polls

The election is coming up. Here's an article that talks about some of the times when polls have been seriously wrong:

http://www.slate.com/id/2271659/

Posted by kshedden at 09:30 PM | Comments (0)

October 19, 2010

Debate on early stopping of clinical trials

If you're interested in clinical trials, here is a juicy debate about something called "early stopping" -- you don't need advanced statistics to understand this.

http://jama.ama-assn.org/cgi/content/full/303/12/1180

http://jama.ama-assn.org/cgi/content/full/304/2/156


http://jama.ama-assn.org/cgi/content/full/304/2/157

http://jama.ama-assn.org/cgi/content/full/304/2/157-a

http://jama.ama-assn.org/cgi/content/full/304/2/158

http://jama.ama-assn.org/cgi/content/full/304/2/158-a

Posted by kshedden at 12:13 PM | Comments (0)

July 22, 2010

Forensic biostatistics

A couple of biostatisticians were the first to identify the flaws in this research that ended up leading to some clinical trials being halted and a medical researcher being suspended from his job:

http:http://prescriptions.blogs.nytimes.com/2010/07/20/duke-suspends-researcher-halts-cancer-studies/?scp=4&sq=cancer&st=cse//

Posted by kshedden at 10:15 AM | Comments (0)

May 06, 2010

Statistical summaries of human development

Here is an interesting discussion about something that comes up frequently -- when a lot of variables are combined to form a summary, the summary may end up tracking almost perfectly with one of the variables. This raises questions about what one really learns by doing this. See the following links for one example.

http://economix.blogs.nytimes.com/2009/05/21/rich-country-within-a-poor-one-part-3/?scp=5&sq=andrew%20gelman&st=cse


http://mapscroll.blogspot.com/search?q=hdi

Posted by kshedden at 05:32 PM | Comments (0)

May 04, 2010

Placebo effect in NYT

A short article on the important issue of placebo responses in medical trials:

http://opinionator.blogs.nytimes.com/2010/05/03/enhancing-the-placebo/?hp

Posted by kshedden at 05:19 PM | Comments (0)

April 28, 2010

Grade Inflation Data

This web site presents some data and analysis relating to grade inflation. As a student, it may interest you to see how UM stacks up compared to other places. As a student of statistics, you may be interested to see how the creators of this web site attempt to convey a somewhat complex analysis in visual form.

http://gradeinflation.com

Posted by kshedden at 09:16 AM | Comments (0)

April 26, 2010

NY TIme article on probability

This is an excellent article:

http://opinionator.blogs.nytimes.com/2010/04/25/chances-are/?hp

Posted by kshedden at 11:05 AM | Comments (0)

April 25, 2010

Job search blog

This seems to provide some good ideas for your present or future job searches:

www.onedayonejob.com

Posted by kshedden at 10:35 PM | Comments (0)

The Case for Undergraduate Statistics

This article is a bit old, but it makes its point very well:

http://www.amstat.org/meetings/jsm/2000/usei/case.html

Posted by kshedden at 10:31 PM | Comments (0)

Revolution computing blog

An R and statistics blog:

http://blog.revolution-computing.com

Posted by kshedden at 10:28 PM | Comments (0)

April 08, 2010

Jared Diamond book on natural experiments

This new book edited by Jared Diamond discusses how empirical methods (mainly statistical methods) are used in some surprising ways:

http://www.hup.harvard.edu/catalog/DIANAT.html

Posted by kshedden at 01:31 PM | Comments (0)

Statistical analysis of ancient languages

An interesting article from the New Scientist:

http://www.newscientist.com/article/dn18725-mathematics-of-ancient-carvings-reveals-lost-language.html

Posted by kshedden at 01:28 PM | Comments (0)

March 23, 2010

NPR Interview: Multiple testing in medical studies

Here is a short interview with a statistician from the National Institute of Statistical Sciences about multiple testing and observational studies:

http://www.npr.org/templates/story/story.php?storyId=124655793

Posted by kshedden at 01:41 PM | Comments (0)

March 19, 2010

Critical article on statistical methods

This article takes a fairly critical view of how statistics is commonly practiced:

http://www.sciencenews.org/view/feature/id/57091/title/Odds_Are,_Its_Wrong

A good chunk of what he says is unfair or misleading. But its worth knowing about this since this viewpoint surfaces from time to time. The stuff about multiple testing is fair and on-target. It's worth a read, and some of the comments are interesting as well.

Posted by kshedden at 04:13 PM | Comments (0)

March 16, 2010

IBM commercials about data analysis

In case you haven't seen these, IBM does a fantastic job advertising statistics ... and you might notice that they never use the word "statistics" -- but that's all they are talking about. Highly recommended.

Why data matters: extracting insights, making better decisions

Business Analytics & Optimization

Identifying Patterns in Data Reveals Insights into the Future


A Smarter Planet Relies On Data Analysis

Intelligent Data Management and Analysis For A Smarter Planet

Making Sense of Data On A Smarter Planet

Data Analysis and Predictions for Smarter Healthcare Decisions:

Smarter Math Builds Equations for a Smarter Planet

Posted by kshedden at 10:59 PM | Comments (0)

March 13, 2010

Articles about data and statistics in The Economist

Here is a series of articles that recently appeared in The Economist magazine about data analysis. You will see that there are opportunities for people with statistics training to have a big impact in all sorts of areas.

Highly recommended reading for anyone who wants to pursue a career involving any kind of analysis or decision making involving data and uncertainty!

The data deluge

All too much (Monstrous amounts of data)

Data, data everywhere

A different game (Information is transforming traditional businesses)

Clicking for gold (How internet companies profit from data on the web)

The open society (Governments are letting in the light)

Show me (New ways of visualizing data)

Needle in a haystack (Uses for information about information)

New rules for big data

Handling the cornucopia

Posted by kshedden at 10:55 PM | Comments (0)

December 18, 2009

Statistics and computational science

This article is about a major new direction in science, in which statistics plays a critical part:

http://www.nytimes.com/2009/12/15/science/15books.html?_r=1&scp=1&sq=data&st=cse

Posted by kshedden at 02:18 PM | Comments (0)

November 24, 2009

Statistics and global development

Some of you may find this video presentation interesting. It uses basic statistics and data visualization to explore the variation in development across countries, and the pace at which countries have developed in history:

http://www.gapminder.org/videos/ted-us-state-department/

Posted by kshedden at 03:48 PM | Comments (0)

November 19, 2009

Statistics articles in NY Times

Here are links to two recent NY Times articles with a lot of statistical
content.

Andy Gelman is a statistician at NYU who has done a lot of work on
statistical models of voter behavior:

http://www.nytimes.com/2009/11/19/opinion/19silver.html?_r=1


Don Berry is a statistician at the M.D. Anderson cancer center who is well
known for his work quantifying the benefits of new cancer screening and
treatment techniques. There are some very interesting issues in the type
of work described here about correcting for "length biased sampling,"
which basically means that when you sample people with a disease, you are
more likely to get people who have a longer, slower progressing form of
the disease than a more rapidly progressing disease. Unfortunately they
don't get into that in the article at all.

http://www.nytimes.com/2009/11/17/health/17cancer.html?ref=health


Posted by kshedden at 09:20 PM | Comments (0)

November 15, 2009

NYT review of Malcolm Gladwell's latest book

There is some interesting statistical content in this book review:

http://www.nytimes.com/2009/11/15/books/review/Pinker-t.html

Posted by kshedden at 11:25 PM | Comments (0)

October 20, 2009

Criticisms of statistical methods in social science

See here for positive and negative views on how statistics is used in social science research:

http://www.nytimes.com/2009/10/20/books/20poli.html

It's not uncommon to hear criticisms that statistical analysis (i) misses the big picture, (ii) is overly concerned with technicalities , or (iii) serves to exclude most people from the discussion. Often these criticisms are true. But sometimes the real issues are statistical and there's no way to make progress other than addressing them head-on.

Kerby

Posted by kshedden at 04:08 PM | Comments (0)

October 12, 2009

NY Times article about analysis of large data sets

All,

This article should help to make clear how many opportunities there are for people with good statistical training (and a little bit of computing knowledge):

http://www.nytimes.com/2009/10/12/technology/12data.html?_r=1&hpw

Kerby

Posted by kshedden at 09:44 AM | Comments (0)

September 23, 2009

Career information for statistics concentrators

The link below is the Career Center's news letter for Statistics concentrators. It has information on the Career Expo which is next week as well as information on internships, jobs, graduate programs and much more. All statistics concentrators are invited and encouraged to meet with a career center advisor and create a Career Center Connector Account (internship/job database).

http://archive.constantcontact.com/fs072/1102219015489/archive/1102719226175.html

http://www.careercenter.umich.edu/students/findingjob/c3student/index.html

Posted by kshedden at 01:31 PM | Comments (0)

September 16, 2009

Occupational outlook for statisticians

This the US Bureau of Labor Statistic's outlook for employment in the statistical field:

http://www.bls.gov/oco/ocos045.htm

Note also this recent article in the NY Times:

http://www.nytimes.com/2009/08/06/technology/06stats.html

Kerby

Posted by kshedden at 03:59 PM | Comments (0)

September 08, 2009

Provost's interview on information technology

Here is an interview with our Provost, Teresa Sullivan, about information technology and research involving large data sets:

http://www.ur.umich.edu/update/archives/090908/sullivan

Posted by kshedden at 09:26 AM | Comments (0)

August 06, 2009

NY Times article about statistics

Here is a link to an article in the New York Times that is very favorable about the future importance of statistics, and opportunities for people in the field:

http://www.nytimes.com/2009/08/06/technology/06stats.html

Posted by kshedden at 10:41 AM | Comments (0)

August 05, 2009

Washington Post article on the Joint Statistical Meetings

The JSM is the main annual meeting of statisticians. This year it was in Washington D.C.

http://www.washingtonpost.com/wp-dyn/content/article/2009/08/04/AR2009080403117.html?hpid=topnews

Posted by kshedden at 06:15 PM | Comments (0)

June 18, 2009

Analytics at IBM

Below are two links to the web site of IBM's Business Analytics and Optimization group. A big part of what they do is statistical:

http://www-935.ibm.com/services/us/gbs/bus/html/bcs_centeroptimization.html?ca=agus_lobbalp-20090518&me=vanity&met=bavan&re=bavan&s_tact=106aw01w&cm_mmc=agus_lobbalp-20090518-106aw01w-_-p-_-bavan-_-bavan


http://www-935.ibm.com/services/us/gbs/bao/ideas.html

Posted by kshedden at 09:58 AM | Comments (0)

May 30, 2009

Statistics in internet advertising

There will be an increasing number of opportunities in the field of computational advertising, as discussed in this article:

http://www.nytimes.com/2009/05/31/business/media/31ad.html

They never use the word "statistics", but that's what it's all about.

Posted by kshedden at 08:01 PM | Comments (0)

April 13, 2009

Some reading on careers

A couple of articles worth reading if you are thinking about possible directions for your future:


http://www.nytimes.com/2009/04/12/weekinreview/12lohr.html


http://www.nytimes.com/2009/04/12/jobs/12starts.html

Posted by kshedden at 02:56 PM | Comments (0)

February 17, 2009

Google executive on statistics

From Hal Varian, Google’s chief economist:

“I keep saying the sexy job in the next ten years will be statisticians. People think I’m joking, but who would’ve guessed that computer engineers would’ve been the sexy job of the 1990s? The ability to take data—to be able to understand it, to process it, to extract value from it, to visualize it, to communicate it—that’s going to be a hugely important skill in the next decades, not only at the professional level but even at the educational level for elementary school kids, for high school kids, for college kids. Because now we really do have essentially free and ubiquitous data. So the complimentary scarce factor is the ability to understand that data and extract value from it.”


The complete article is here:

http://www.mckinseyquarterly.com/Strategy/Innovation/Hal_Varian_on_how_the_Web_challenges_managers_2286

Posted by kshedden at 09:42 AM | Comments (0)

January 14, 2009

R in the news

Those of you who love or hate R may be interested in this:

http://www.nytimes.com/2009/01/07/technology/business-computing/07program.html?em

Posted by kshedden at 08:33 PM | Comments (0)

January 08, 2009

Employment for statisticians

Those of you worried about the economy and your future employability may be reassured somewhat by these articles. The second one doesn't mention statistics explicitly but the industries that are referred to hire a lot of statisticians.

http://online.wsj.com/article/SB123119236117055127.html?mod=rss_Today%27s_Mo
st_Popular

http://online.wsj.com/article/SB122999321221628707.html

Follow-up (added 1/12): check this out for an alternative perspective:

http://www.thebigmoney.com/articles/hey-wait-minute/2009/01/09/im-lumberjack-and-thats-ok

Posted by kshedden at 08:30 PM | Comments (0)