The American Statistician Highlights
Leading Papers Look at Teaching Bayes
to Nonstatisticians
Stigler Talks Harvard Statistics in
History Corner
Peter Westfall, The American Statistician Editor
and teach Bayes to social scientists. The theme
books Reviewed
of this section, “Teaching Bayesian Statistics
to Nonstatistics Graduate Students,” will be
Handbook of Univariate
continued in an invited session of the Joint
and Multivariate Data Analysis
Statistical Meetings in Washington, DC, in
and Interpretation With SPSS
August 2009.
by Robert Ho
I am also happy to highlight a paper in
Introduction to Bayesian Statistics
the History Corner by Stephen Stigler,
(2nd ed.) by William M. Bolstad
titled “Statistics at Harvard? A Toast to the
Statistics Department,” which is an infor-
Introduction to Bioinformatics by Anna
Benjamin Peirce (1809–1880) lecturing,
mative and often amusing account of the
Tramontano
possibly on probability. From “A Toast to the
formative years of statistics (or not, as the
Introduction to Business Statistics (6th ed.)
Harvard Statistics Department,” by Stephen
case may have been). The account is com-
M. Stigler, featured in the August 2008 issue
by Ronald M. Weiers
plete with historical photos of some of
of The American Statistician.
the individuals, as well as a toast to their
Missing Data: A Gentle Introduction
by Patrick E. McKnight, Katherine M.
M
any in the teaching profession have current excellence. Ronald Snee also contrib-
McKnight, Souraya Sidani, and Aurelio
felt the need to provide more train- utes to the History Corner with a look back
José Figueredo
ing in Bayesian statistics to gradu- at W. Edwards Deming and a look forward to
ate students who are not statisticians, as their “lean Six Sigma.” Modern Engineering Statistics by Thomas
academic literature has become increasingly Can Bayesian methods correct for selec- P. Ryan
infused with Bayesian methods. Much of this tion effects? Stephen Senn addresses this
Multidimensional Nonlinear Descriptive
impetus comes from the graduate students, issue in the lead article of the General sec-
Analysis by Shizuhiko Nishisato
themselves, who wish to participate in the tion, with an answer veering toward “no.”
publication process using Bayes for their How do sufficient statistics look when the
Probability and Statistics for Computer
research, or at least wish to understand their parameter space is finite? Lynn LaMotte
Scientists by Michael Baron
literature more clearly. This is a difficult tells us. How can the Moore-Penrose
The R Book by Michael J. Crawley
group. Their mathematical training is usually inverse be characterized? See Xiaomi Hu’s
not great, and they are not inclined to take paper. Does E(Y
n
|X)= Y
n
and E(X
n
|Y)= Speckman, Jeffrey Rouder, Richard Morey,
more classes in basic math—or any nonsub- X
n
, all n imply X and Y are independent? and Michael Pratte then show how to visual-
ject matter. The answer is no; Áureo de Paula provides a ize stochastic order by using the “delta plot.”
A series of four papers leads the August counterexample. Coming full circle, we have our usual
issue of TAS, aimed at providing help for edu- Moving to the Statistical Practice section, Teacher’s Corner articles. This issue features
cators wishing to teach this audience, within Deborah Peikes, Lorenzo Moreno, and the use of “real-world” HIV infection data
the existing constraints. The first, by Byron Sean Michael Orzol compare propensity in a paper by Chad Bhatti and Jennifer
Gajewski and Stephen Simon, proposes score matching with randomized compara- Wightman to make Bayes theorem more
a “one-hour seminar” for teaching Bayes to tive experiments for evaluating social pro- relevant to students. The section concludes
nursing graduate students. The second, by grams and identify bias in the propensity with a discussion by Duncan Murdoch and
Greg Allenby and Peter Rossi, discusses score matching method. Rajdeep Grewal, Yu-Ling Tsai about teaching that p-values are
teaching a full-semester course to graduate James Dearden, and Gary Lilien analyze random variables to avoid misinterpretation.
students in business. The third, by Wesley data on everyone’s favorite topic—university Finally, there is a rather interesting and
Johnson and Jessica Utts, discusses course rankings—finding a great deal of inertia. unusual letter to the editor by Hubert
options, including seminars and full semes- In the Statistical Computing Graphics Wong, who comments critically on a paper
ters, often with a graduate veterinary audience section, David Scott and Warren Scott by Don Rubin that appeared in JASA in
in mind. Andrew Gelman’s paper rounds out develop a simple, yet efficient, smoothed his- 2005 about direct causal effects. Rubin
the series with discussion of how to motivate togram when the intervals are irregular. Paul provides a response. n
16 AMSTAT NEWS AUGUST 2008
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68