Notes for Friedman, Pisani and Purves

• Recommended supplementary reading:
  1. The Cartoon Guide to Statistics by Larry Gonick and Woollcott Smith.
  2. The Manga Guide to Statistics by Shin Takahashi. I haven't seen this one myself, but reviews (from professors as well as students) are very favorable.
  3. Drunkard's Walk: How Randomness Rules our Lives by Leonard Mlodinow.
  4. The Flaw of Averages: Why we Underestimate Risk in the Face of Uncertainty by Sam L. Savage. A very readable book on the most common errors made in using statistics. Focus on business applications, as the author has been a consultant to many firms.

• Statistics and the Financial Meltdown
  • Risk Mismanagement by Joe Nocera, New York Times, 4 Jan. 2009. Despite the criticisms below, it is a good article to start with.
  • Woefully Misleading Piece on Value at Risk in the New York Times by Yves Smith. Smith thinks the article above is misleading (I would be gentler and just call it incomplete) because insufficiently aware of the statistical assumptions of the Value-at-Risk models. She points out that it is well known that the normal distribution is well known to model uncertainty inaccurately, and that a proper model should include skewness and kurtosis .
  • Risk Management for Beginners by James Kwak. Adds the perhaps more important criticism that VaR models do not model risk at all. See the comments for the distinction between uncertainty and risk.
  • Medical trials in the news
    • For Widely Used Drug, Question of Usefulness Lingers, by Alex Berenson. Sept. 1, 2008 issue of the New York Times
    • Drug Companies and Doctors: A Story of Corruption by Marcia Angell, former editor of the New England Journal of Medicine, reviewing three recent books.
      New York Review of Books, 56:1 (Jan. 15, 2009).

  • Simpson's Paradox:
    • Job training example by Jeff Gill
    • Baseball averages of Derek Jeter and David Justice show Simpson's paradox.
    • Red State, Blue State: What's the Matter with Connecticut by Andrew Gelman. Simpson's paradox in voting behavior.
    • Prosperity and Views about the European Union European Union popular among richer voters in poorer countries.

    ---

  Histograms and data

  • Educational Attainment in the US: 2007 From the Census Bureau's Current Population Survey.
  • Table2 for those 25 and older is the most convenient for class work.
  • Household Income in 2006
    Based on the Census Bureau's Current Population Survey, March 2007. It would be a good exercise to see if you can take the given percentages, calculate the densities, and draw the histogram.

  • Random Number Generation from Wikipedia.
  • Mersennne Primes
  • Toss Out the Toss-Up: Bias in Heads-or-Tails Erica Klarreich, Science News, 165:9 (Feb. 28, 2004), 131.
  • Dynamical Bias in the Coin Toss by Persi Diaconis, Susan Holmes, and Richard Montgomery. SIAM Review, 49:2, 211-235.

  Psychology of Probability

  Link to Kevin Drum's Chart of the Day for Nov. 27, 2008 by clicking on the picture. 1,540 adults were asked how many times a coin would turn up heads. Their average response was of course 5 -- until the next question, asking them how often it would happen if they were to win 10 euros each time it did.

  The average response then became 3.9 heads -- most people seem to think they are less lucky than average.

  Drum notes that 10 or so people thought they would win all ten times, and comments "Here in America, we call those people 'investment bankers' ".

  Elyes Jouini, the best-known author of the paper, is a finance professor at the University of Paris; attitudes to risk are central to understanding the operation of financial markets.