prove that any permuted column sampling plan that eliminates
estimation bias, using the smallest possible number of runs in each
The lead article in the May 2008 issue of
array and containing the largest possible number of arrays, can be
Technometrics is a panel discussion about the future
characterized by an orthogonal array of strength 2. Approximate
of industrial statistics. The editors and panelists see
standard errors of the first-order sensitivity indexes are derived for
the article not as the end of the story, but the take-
this sampling plan. Two examples are given demonstrating the
off point for further discussion. To that end, they
sampling plan, behavior of the estimates and standard errors, and
are initiating an open discussion forum. You can
comparative results based on other approaches.
take part by going to www.asq.org/pub/techno and
Jian Zhang, Peter F. Craigmile, and Noel Cressie consider
clicking on Networking and Events.
a common problem in spatial statistics—estimation of a spatial
exceedance region—in their article, “Loss Function Approaches To
Predict a Spatial Quantile and Its Exceedance Region.” A spatial
exceedance region is the set of points at which a process is larger
type of experiment separately. The modeling approach exploits than some specified quantile. Such regions are scientifically impor-
Bayesian hierarchical Gaussian process models. The heterogeneity tant because of their ability to reflect unusual events with strong
among data sources is accounted for by performing flexible location environmental impact. The authors use classes of loss functions
and scale adjustments. The approach tends to produce predictions based on image metrics to predict spatial exceedance regions and
closer to those from the high-accuracy experiment. The proposed propose a joint loss to predict a spatial quantile and its exceedance
method is illustrated with two examples, one with detailed and region. The optimal predictor is obtained by minimizing the pos-
approximate finite elements simulations for mechanical material terior expected loss given the process parameters, which is achieved
design and the other with physical and computer experiments for by simulated annealing. Various predictors are compared through
modeling a food processor. simulation. The methodology is applied to a spatial data set of tem-
Computer models are often used for carrying out sensitiv- perature change over the Americas.
ity analysis. Max D. Morris, Leslie M. Moore, and Michael D. The issue closes with a short article by Georges H. Guirguis,
McKay develop ideas for the design of such studies in their article, titled “A Note on Computing the Probability and Critical Values
“Using Orthogonal Arrays in the Sensitivity Analysis of Computer for the Half-Normal Plot.” The half-normal plot is one of the
Models.” The article considers a class of input sampling plans called standard tools used in analyzing factorial and fractional-factorial
permuted column sampling plans. These plans, which are popu- experiments. Zahn showed how to add “guardrails” to these plots,
lar in sensitivity analysis of computer models, support estimation for use in deciding whether contrasts differed significantly from
of first-order sensitivity coefficients, but lead to biased estimates zero. Guirguis shows the calculation of the critical values for those
when the usual practice of random column permutation is used tests can be reduced to a one-dimensional integral, providing much
to construct the sampling arrays. Deterministic column permuta- faster computation of the critical values than with Zahn’s formulas,
tions may be used to eliminate this estimation bias. The authors which required a two-dimensional numerical integration.
■
Second Workshop for Program Chairs Scheduled
O
n August 1 and 2, the Friday department chairs and with the admin- participant to help cover the cost of
and Saturday prior to JSM, the istration), administrative activities (e.g., travel for the speakers and for food.
Caucus of Academic faculty reviews, budgets), and working (This is in addition to and separate
Representatives will hold the second with the department’s administrative from JSM registration). Registration
annual workshop for chairs of programs staff. For more experienced chairs, there for the Saturday sessions only is $150,
in statistics and biostatistics. This two- will be sessions on interactions outside which includes the Friday dinner.
day workshop will provide information the university, department growth and Hotels should be booked through the
and discussion for leaders of statistics the student pipeline, and faculty issues JSM meeting site. Registration for the
and biostatistics groups and depart- (e.g., diversity, two-body problem). The workshop is available online at www.
ments. In particular, it will stimulate workshop is designed so the first day amstat.org/meetings/cpsb. Space is lim-
discussion between new and experienced is primarily for new and less experi- ited and will be filled on a first-come,
chairs on a range of topics as broad as enced chairs, while the second day is first-served basis.
development of long-range planning for all chairs. Dinner on Friday is for Program committee members for this
goals and as specific as recommenda- all registrants. workshop are Doug Simpson, Dalene
tions for dealing with the media. To keep registration fees reason- Stangl, and Ken Koehler. The ASA
Agenda items of particular interest able, the caucus is providing par- liaison is Keith Crank. Please address
to new chairs include intra-university tial support for the workshop. There questions to dgs@uiuc.edu or keith@
collaborations (e.g., working with other will be a registration fee of $250 per amstat.org.
20 AMSTAT NEWS MAY 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  |  Page 69  |  Page 70  |  Page 71  |  Page 72  |  Page 73  |  Page 74  |  Page 75  |  Page 76  |  Page 77  |  Page 78  |  Page 79  |  Page 80  |  Page 81  |  Page 82  |  Page 83  |  Page 84  |  Page 85  |  Page 86  |  Page 87  |  Page 88  |  Page 89  |  Page 90  |  Page 91  |  Page 92