National
Institute of Statistical Sciences
19 T. W. Alexander
Drive
P.O. Box 14006
Research Triangle Park, NC 27709-4006
Tel: 919.685.9300
FAX: 919.685.9310
admin@niss.org
Report of the Director
The year began on November 7, 1997, with the dedication of the NISS building as the opening event of the Annual Meeting of the NISS Corporation. Nearly 200 people attended the ceremonies and reception, including Dennis Wicker, Lieutenant Governor of North Carolina. Sherwood Smith, Chair of the Board of Directors of TUCASI (the Triangle Universities Center for Advanced Studies), acted as master of ceremonies, and introduced Mr. Wicker, who gave the opening address. Other speakers were Jon Kettenring, John Bailar and Jerome Sacks.
The primary focus of the Annual Meeting was on the submission to the National Science Foundation competition for Mathematical Sciences Research Institutes. A proposal was developed and submitted in January, 1998. Five finalists, for an unspecified number of awards, were selected in July, 1998. Site visits were held during September and October, 1998; NISS was visited in September. Final selections are expected by the end of the year. There was considerable help from a number of Trustees and Members in preparing and participating in the site visit.
A principal operation during the year was continued research on transportation, including work funded by the Los Alamos National Laboratory on synthesis of regional travel activity. Other major activities included continued research on software engineering (in collaboration with Lucent Technologies), developing protocols for use and dissemination of EPA data sets, continued work on drug discovery, study of network intrusion (in collaboration with AT&T), beginning work on bioinformatics, and recruitment of new Junior Fellows.
New proposals were generated, most notably to initiate efforts in information technology. Workshops were held on NISS transportation research and materials science. NISS also held a workshop to help disseminate information about a new NSF-wide program in Knowledge and Distributed Intelligence. Presentations of NISS research occurred at national and international conferences, and NISS personnel organized sessions at national and regional meetings of several societies. NISS co-sponsored KDD 98 (Knowledge Discovery and Data Mining), a conference devoted to approaches for dealing with large data sets.
Measurement, Modeling and Prediction for Infrastructural Systems. This five-year, NSF-funded project began on November 1, 1994. Principal investigators are A. Karr (NISS), N. Rouphail (NCSU) and J. Sacks (NISS). (E. Pas (Duke) was one of the original PI's; his untimely death in November, 1997, led to the designation of Rouphail as co-PI.)
Research on travel demand forecasting has concentrated on synthesizing populations from aggregated data and multinomial choice modeling. The principal participants were C. Bhat (Texas), F. Koppelman (NU), P. Koskenoja (NISS), P. Speckman (Missouri) and D. Sun (Missouri). Aspects of this work overlap with work on the LANL project (see below).
Work on network modeling has addressed equilibria of large network models and the validation of such models. The key figure here is D. Boyce (UIC) with the assistance of P. Nelson (UIC), R. Easch (CATS), R. Buck (Western Michigan) and Sacks. Effective code has been developed and computer experiments are being conducted to improve the running time of the code. The focus in the coming year will be on developing a validation process for the model - identifying data which can be used to test model predictions with appropriate uncertainties.
Work on vehicle matching, visualization of freeway travel times and assessment of algorithms for incident detection have been the focus of work at UC-Berkeley. The work has involved P. Bickel and J. Rice (California--Berkeley), Y. Ritov (Hebrew University) and K. Petty (NISS) with input by Sacks. Two presentations by thesis students were made at the Joint Statistical Meetings in Dallas in August, 1998.
A Phase I study of the microsimulator (CORSIM) was completed. CORSIM is to be used as a platform for evaluating signalization strategies. Plans are underway and data are being gathered to produce a (Phase II) signalization study on a complex network. The Phase I study collected data and identified behavior of CORSIM on a small network of the Chicago Business District. This was done with the cooperation of the Chicago Department of Transportation (CDOT) and involved Rouphail, A. Sen (UIC, who was sworn in as Director of the U.S. Bureau of Transportation Statistics on November 4, 1998), V. Thakuriah (UIC), S. Stidham (UNC--CH), E. Miller--Hooks (NISS), Karr and Sacks. Phase II likewise involves CDOT, as well as B. Park (NISS), a recently appointed Junior Fellow, who also played an important role in bringing Phase I to a conclusion.
Work on materials science and deterioration of concrete, primarily at the Center for Advanced Cement Based Materials at Northwestern University (NU), has focused experiments for measuring and models for predicting water permeability of cracked samples of concrete from materials variables and characteristics of cracks. A new program of numerical experimentation, to simulate permeability and damage, was initiated. This work is led by Karr and S. Shah (NU); key participants are B. Ankenman (NU), T. Igusa (NU), J. Picka (NISS, located at NU) and C. Aldea (NU).
Statistically-Based Activity Generation. Substantial progress was made in the first year of a two-year contract with Los Alamos National Laboratory (LANL). The goal is to perform the research necessary to provide regional-scale activity patterns as inputs to the LANL-developed TRANSIMS model, to model travel and traffic in Portland, OR. Models and algorithms have been devised and implemented to produce a first set travel--activity patterns for the population in the Portland metropolitan area. The project is directed by Karr in cooperation with R. Beckman (LANL). Key participants are Speckman and Sun (Sun is visiting NISS from September 1998, to June, 1999) with added participation of Kopppelman, K. Vaughn (Metropolitan Transportation Commission, Oakland, CA) and a recently appointed NISS Junior Fellow, J. Lee.
ITS Integration of Real-Time Emissions Data and Traffic Management Systems. This project, funded by the ITS/IDEAS program of the Transportation Research Board, has collected field data that relate pollutant emissions to traffic characteristics at various levels of integration. Significant instrumentation and measurement issues had to be addressed; data are now being analyzed. This study is run by Karr and Rouphail with important involvement of C. Frey (NCSU) and C. Gu (Purdue).
Indices of Environmental Status and Trend. As part of a consortium (which includes American, George Mason, Maryland--Baltimore County and Penn State) organized by the EPA's new Center of Environmental Information and Statistics, NISS has worked on protocols and case studies for the public release and use of environmental data. Sacks has been leading this effort. A study on ozone trends in the Raleigh-Durham area was completed by a research assistant at NCSU under the direction of P. Bloomfield (NCSU). Another study of safety of drinking water was completed by J. Hilden-Minton (NISS), with advice from R. Smith (UNC-CH). A third study is underway on particulates in Forsythe County, North Carolina.
Statistical Research and Analysis for Technical Services: CASTNET. A one-year contract with EPA focused on studying CASTNET, a monitoring network chiefly involved with the pollutant SO2 (acid deposition). This work was successfully completed with reports and spinoff papers. The issues involved space-time modeling of SO2 over the eastern U.S. The work was done in collaboration with EPA researchers L. Cox and D. Holland, was led by Sacks and Smith, and had major involvement of V. De Oliveira and N. Saltzman (NISS). The spinoff work on selecting prior distributions for correlation parameters involved J. Berger (Duke).
Code Decay in Legacy Software Systems: Measurement, Modeling and Statistical Strategies. This partnership with Lucent Technologies is in its last of three years. Co-project directors are Karr and S. Eick of Lucent. The project is directed at improving the development process for large legacy software systems by application of statistical strategies to model and control the effects of design decisions, software architecture and organizational factors. Key participants are T. Graves (NISS), J. S. Marron (UNC--CH), A. Mockus (Lucent), A. Porter (Maryland), N. Staudenmayer (Duke) and L. Votta (Lucent). Principal products of the research during the past year were the first new network models for the breakdown of modularity over time, methods to impute change effort that provide additional evidence of decay and can be used to evaluate effectiveness of tools used by developers, and a major organizational study.
The emphasis in the third year will be on ``perfective maintenance:'' is re-engineering software in anticipation of future changes cost-effective?
Pilot Projects to Explore Large Data Sets. This research effort comprises two interconnected pilot projects dealing with large data sets, each involving a major industrial partner. In both instances, there are specific scientific issues with high-stakes implications for the industry at large; the available data sets are for current and prospective needs, lending immediacy to the research. At the same time, the pilot projects address long-run issues (such as heterogeneity and multiplicity) whose relevance extends beyond the corporations and industries they involve.
Drug Design. The component on drug discovery is developing sequential strategies to enable efficient High Throughput Screening of drug compounds for chemical activity. Results have been presented at a variety of meetings (including the 1998 Interface); reports have been completed on general sequential strategies, multiple classification trees and pooling. This project is being carried on in close collaboration with Glaxo Wellcome, Research Triangle Park, NC. The effort is led by Sacks and S. Young (GW). Key researchers involved were Gu, Y.-B. Lim (NISS visitor from Korea), A. Menius (GW), A. Rusinko (GW), K. Tatsuoka (NISS) and M. Xie (NISS, Rutgers).
Work has also been started on extending notions developed in drug discovery to problems in bioinformatics, especially on gene discovery and expression as well as in the modeling of protein expression. This has involved Young, Tatsuoka, A. Stark (NISS), F. Seillier-Moiseiwitsch (UNC--CH). Other related work is just beginning, with the involvement of B. Weir (NCSU) and J. Graham (NISS).
Network Intrusion. The first year of this effort produced strategies for identifying network intruders by modeling user profiles and identifying anomalous patterns. Current and next year's efforts are to model potential customer "defections" through use of network data - producing statistical strategies for capitalizing on large network data to predict individual user behavior. Directing this research are Karr and D. Pregibon (AT&T). Key participants are M. Schonlau (NISS), Y. Vardi (Rutgers), W. DuMouchel (AT&T), R. Bell (RAND Corporation, visiting AT&T) and N. Raghavan (Ohio State, visiting NISS).
Education Studies. A contract with the National Center for Educational Statistics (NCES) for a study about reporting analyses for small subgroups of a population was completed. The report was drafted by L. Jones (UNC-CH), with the assistance of J. Blair (Maryland), J. Chromy (RTI), R. Jaeger (UNC-G), B. Spencer (NU), D. Thissen (UNC-CH) and Sacks. A follow on study by Spencer, with support by Jones, has resulted in a preliminary study currently being circulated for comment.
A second NCES-funded study on how to treat missing data has resulted in a preliminary report drafted primarily by Chromy with contributions from a panel that included Jones, Olkin, Blair, Jaeger, G. Kalton (Westat), R. Little (Michigan) and V. Williams (NCCU). This report stemmed from a two-day meeting held at NISS, which included a number of officials and researchers from NCES.
North Carolina School of Science and Mathematics. NISS developed a framework, strategy and initial database for evaluation of NCSSM's Educational Futures Center (EFC). The EFC mission is to create a state-wide network of "cyber-campuses" that will carry NCSSM's programs beyond its residential campus in Durham, NC. Innovative electronic and educational technologies are involved. Leading the effort were J. Hattie (UNC-G) and Jaeger; Karr and Sacks have been involved.
NCSSM has recently received a 5-year, $6,500,000 grant from the Office of Education with NISS as the evaluation arm of the project. The evaluation will implement the previously devised strategy and extend it to deal with teacher development.
Access and Confidentiality of Federal Data Sets. NISS has pending at the NSF a proposal to develop a (prototype) World Wide Web-based system to allow adequate access to confidential data from Federal agencies while maintaining low risk of disclosure. The proposal has Karr as PI, with S. Keller-McNulty (LANL) as co-PI. It involves MCNC (a major network facility in Research Triangle Park, adjacent to NISS), S. Fienberg, G. Duncan and L. Sweeney (Carnegie-Mellon), M. Franklin (Maryland) and A. Saalfeld (Ohio State), as well as several Federal agencies: the Bureau of the Census, Bureau of Labor Statistics, National Center for Health Statistics and the National Agricultural Statistics Service.
Knowledge in Virtual Organizations. This proposal was developed to understand how knowledge is transmitted, monitored and evaluated within virtual (network-based) organizations. This was to be done in cooperation with an assortment of researchers from Lucent, DePaul, Illinois, UNC-CH and Duke. The proposal was not funded, but several aspects will be pursued as part of an overall NISS initiative in information technology.
Junior Fellows are generally supported partly by funds from NSF and partly by funds from NISS projects. They work on one or more projects that match their backgrounds and career goals, with time available as well to pursue independent research activities. The Advisory Committee is composed of P. Bickel, S. Fienberg, S. Geman, G. McDonald, G. Wahba and S. Zeger.
Fellows whose appointments at NISS ended during the past year are
The status of continuing appointees is as follows:
New appointees in 1998 are:
Computing. A second dual-processor HP server was added to NISS' computing facilities during the summer of 1998. Charles Hooks, Computational Systems Manager, left NISS to take up a position at Penn State University. He was replaced by Lael Tucker, who has maintained and improved the NISS system.
Publications. Plans are underway to reduce and eliminate hard-copy distribution of NISS technical reports, replacing them with Web-accessed documents.
Visitors and other Appointments. Visits to NISS were made by D. Daley (Australian National), working on the transportation project, C. Gu (Purdue), working on emissions and drug design and Y.-B. Lim (Korea) working on drug discovery.
Current visitors are J. Lynch (South Carolina) working on materials and leading a seminar to develop a potential project, D. Sun (Missouri) working on transportation projects, N. Raghavan (Ohio State) working on network data and materials.
B. McNeney has an appointment at NCSU, partially supported at NISS, to work on problems in clinical trials while mentored by A. Tsiatis (NCSU). McNeney is a recent Ph.D. (Biostatistics, Washington). A research associate appointment has been made for Hilden-Minton to assist on environmental work and on Digital Government efforts. A research associate appointment is being planned for Pan-Wei Cheng (Ph.D., Economics, Illinois-Chicago) to work on transportation and commodity flows.
October 26, 1998
 |
 |
| Help |