Serious links

• Monte Carlo and quasi-Monte Carlo wiki
• Guaranteed Automatic Integration Library (GAIL). Different from gail.com
• Andrew Gelman's blog
• Christian Robert's blog
• Geoffrey Mess, brilliant man left us too soon.
• My friends at Google Consumer Surveys did well in the 2012 election
• Jeffrey Rosenthal's home page
• Alan Genz's home page
• Radford Neal's home page
• Fred Hickernell's home page.
• Pierre L'Ecuyer's home page.

Fun links

• Scribing a class lecture, the grad student desperately attempts to create something of value
• In honour of Ian Sloan's 70th birthday
• Roger Koenker's undata collection
• Stephen Leacock (approximately the Canadian Mark Twain) has a fun story about a swimming teacher who cannot swim. There are several versions of it out there. There is a mellow one here under "How Teachers Swim'' and a somewhat edgier one here.

Coffee

• NCA one (30 sec.)
• NCA two (30 sec.)
• CGP Grey (4 min, 20sec, might serve an ad)

Some scholarly thoughts that resonate (YMMV)

• Eric Gilbert's PhD Syllabus Good comments on how advisor/advisee should work together. Designed for CS but largely relevant for statistics.
• Matthias Schonlau's advice on how to finish a Ph.D. and more
  Some people struggle a bit, some a lot, some hardly at all. Others might not know your situation.
• Gopen and Swan on the science of scientific writing
• Richard Hamming on you and your research
• Not a link, but I found the quote below in an article that I got from Wayne Oldford. The article was by George Barnard from around 1989, when he was at Essex. I can't find it on the web but it was entitled "Some remarks on the elementary theory of tests of significance''. Barnard is writing about V. P. Godambe:

  Godambe, better than anyone else I know, appreciates the point often made to me by R. A. Fisher, that in statistical inference, as distinct from mathematical inference, there is a world of difference between the two statements "p is true" and "p is known to be true".

  It's really just about one more level of indirection but it can make a difference about whether a proposed solution to a problem can actually be used.