Showing papers in "Statistical Science in 1997"
TL;DR: In this article, the authors consider Fisher's changing justifications for the method, the concepts he developed around it, and the approaches he discarded (including inverse probability) and the approach he discarded.
Abstract: In 1922 R. A. Fisher introduced the method of maximum likelihood. He first presented the numerical procedure in 1912. This paper considers Fisher's changing justifications for the method, the concepts he developed around it (including likelihood, sufficiency, efficiency and information) and the approaches he discarded (including inverse probability).
622 citations
TL;DR: In this article, interior point methods were combined with a new statistical preprocessing approach for quantile regression in order to obtain a 10- to 100-fold improvement in computational speeds over current (simplex-based) $\ell_1$-algorithms in large problems.
Abstract: Since the time of Gauss, it has been generally accepted that $\ell_2$-methods of combining observations by minimizing sums of squared errors have significant computational advantages over earlier $\ell_1$-methods based on minimization of absolute errors advocated by Boscovich, Laplace and others. However, $\ell_1$-methods are known to have significant robustness advantages over $\ell_2$-methods in many applications, and related quantile regression methods provide a useful, complementary approach to classical least-squares estimation of statistical models. Combining recent advances in interior point methods for solving linear programs with a new statistical preprocessing approach for $\ell_1$-type problems, we obtain a 10- to 100-fold improvement in computational speeds over current (simplex-based) $\ell_1$-algorithms in large problems, demonstrating that $\ell_1$-methods can be made competitive with $\ell_2$-methods in terms of computational speed throughout the entire range of problem sizes. Formal complexity results suggest that $\ell_1$-regression can be made faster than least-squares regression for n sufficiently large and p modest.
549 citations
TL;DR: The authors showed that the conditional frequentist method can be made virtually equiva- lent to Bayesian testing, which is of considerable interest because it is often perceived that Bayesian and frequentist testing are incompatible in this situation.
Abstract: In this paper, we show that the conditional frequentist method of testing a precise hypothesis can be made virtually equiva- lent to Bayesian testing. The conditioning strategy proposed by Berger, Brown and Wolpert in 1994, for the simple versus simple case, is gener- alized to testing a precise null hypothesis versus a composite alternative hypothesis. Using this strategy, both the conditional frequentist and the Bayesian will report the same error probabilities upon rejecting or ac- cepting. This is of considerable interest because it is often perceived that Bayesian and frequentist testing are incompatible in this situation. That they are compatible, when conditional frequentist testing is allowed, is a strong indication that the \wrong" frequentist tests are currently be- ing used for postexperimental assessment of accuracy. The new unied testing procedure is discussed and illustrated in several common testing situations.
182 citations
TL;DR: Key contributions in the area of statistics as applied to the use of molecular marker technology and quantitative genetics in the search for genes affecting quantitative traits responsible for specic diseases and economically important agronomic traits are reviewed.
Abstract: This article reviews key contributions in the area of statistics as applied to the use of molecular marker technology and quantitative genetics in the search for genes affecting quantitative traits responsible for specic diseases and economically important agronomic traits. Since an exhaustive literature review is not possible, the limited scope of this work is to encourage further statistical work in this vast eld by rst reviewing human and domestic species literature, and then concentrat- ing on the statistical developments for experimental breeding popula- tions. Substantial gains have been made over the years by both plant and animal breeders toward a long-term goal of locating genes affect- ing quantitative traits (quantitative trait loci, QTLs) for the eventual characterization and manipulation of these genes in order to develop im- proved agronomically important traits. Our main concern is that the care and expense that are required in generating both genetic marker data and quantitative trait data should be accompanied by equal care in the statistical analysis of the data. Through an example using an F 2 male genetic map of mouse chromosome 10, and quantitative trait values mea- sured on weight gain, we implement much of the reviewed methodology for the purpose of detecting or locating a QTL having an effect on weight gain.
168 citations
TL;DR: In this article, the authors review the properties of hazard models explicitly representing the effects of unobserved, and partially observed, stochastic covariates. And they show that marginal survival distributions and likelihoods generated in analytically closed form make such parametrically detailed models computationally tractable.
Abstract: Stochastically changing covariates may influence survival. They may be observed, unobserved or partly observed. We review the properties of hazard models explicitly representing the effects of unobserved, and partially observed, stochastic covariates. Such models will increase in importance as new longitudinal population studies, and longitudinal surveys of high dimensional failure processes in humans, become available--many are now in progress. It is shown that marginal survival distributions and likelihoods generated in analytically closed form make such parametrically detailed models computationally tractable. Several ways of defining the marginal distribution of the data for constructing a likelihood function are considered. The most complete models can handle both continuously and discretely evolving covariates. Parameters can be estimated from multiple data sets to retrospectively and prospectively evaluate covariate trajectories. Such methods will both extract more information from a longitudinal study and use it in a parametric structure that is logically consistent with the behavior of the underlying processes of substantive interest.
132 citations
109 citations
Journal Article•
106 citations
TL;DR: A Bayesian approach is introduced which estimates and adjusts for publication bias in the passive smoking meta-analysis, and it is estimated that the estimated excess risk may be overstated by around 30%, both in U.S. studies and in the global collection of studies.
Abstract: "Publication bias" is a relatively new statistical phenomenon that only arises when one attempts through a meta-analysis to review all studies, significant or insignificant, in order to provide a total perspective on a particular issue. This has recently received some notoriety as an issue in the evaluation of the relative risk of lung cancer associated with passive smoking, following legal challenges to a 1992 Environmental Protection Agency analysis which concluded that such exposure is associated with significant excess risk of lung cancer. We introduce a Bayesian approach which estimates and adjusts for publication bias. Estimation is based on a data-augmentation principle within a hierarchical model, and the number and outcomes of unobserved studies are simulated using Gibbs sampling methods. This technique yields a quantitative adjustment for the passive smoking meta-analysis. We estimate that there may be both negative and positive but insignificant studies omitted, and that failing to allow for these would mean that the estimated excess risk may be overstated by around 30%, both in U.S. studies and in the global collection of studies.
104 citations
TL;DR: In this article, a Bayesian latent trait formulation was proposed as a replace- ment for GPA-based class ranks at Duke University to eliminate many of the inequities as- sociated with GPA based measures.
Abstract: In response to the growing problem of grade ination in Amer- ican undergraduate institutions, alternatives to GPA and GPA-based stu- dent assessment are discussed One alternative summary, based on a Bayesian latent trait formulation, eliminates many of the inequities as- sociated with GPA-based measures and has been proposed as a replace- ment for GPA-based class ranks at Duke University 1 BACKGROUND Grade point average, or GPA, is the most widely used summary of undergraduate student perfor- mance in our educational system Unfortunately, combining student grades through simple averag- ing schemes to obtain GPA's results in systematic biases against students enrolled in more rigorous curricula and has important consequences in stu- dent course selection It creates perverse incentives for faculty to inate grades and lower standards, and it rewards students for selecting less chal- lenging courses and majors (Larkey and Caulkin, 1992)
80 citations
Journal Article•
53 citations
TL;DR: The method of maximum likelihood was introduced by R. A. Fisher in 1912, but not until 1922 under that name as mentioned in this paper, and the phrase "inverse probability" was used in various ways before defining "likelihood" in 1921 to clarify his meaning.
Abstract: The method of maximum likelihood was introduced by R. A. Fisher in 1912, but not until 1922 under that name. This paper seeks to elucidate what Fisher understood by the phrase "inverse probability," which he used in various ways before defining "likelihood" in 1921 to clarify his meaning.
TL;DR: Doob was a member of the faculty of the University of Illinois from 1935 until his retirement in 1978 and was Commissar of the Champaign-Urbana Saturday Hike for about 25 years after World War II as mentioned in this paper.
Abstract: Joseph L Doob was born in Cincinnati, Ohio, February 27, 1910 He received the degrees AB in 1930, AM in 1931 and PhD in 1932 from Harvard University From 1932 to 1934 Doob did postdoctoral work at Columbia University In 1933-1934 he held a Carnegie Corpo- ration Fellowship Doob was a member of the faculty of the University of Illinois from 1935 until his retirement in 1978 He was Commissar of the Champaign-Urbana Saturday Hike for about 25 years after World War II Doob is a member of the National Academy of Sciences and Foreign Associate of the Academy of Sciences, France He was President of the Institute of Mathematical Statistics in 1950 and of the American Math- ematical Society in 1963 and 1964 He received the National Medal of Science in 1979 In 1984 Doob was awarded the Career Prize by the American Mathematical Society for "his fundamental work in establish- ing probability as a branch of mathematics and for his continuing pro- found influence on its development"
TL;DR: In this paper, the authors present an account of the life of the author's book Testing Statistical Hypotheses, its genesis, philosophy, reception and publishing history, and some discussion of the position of hypothesis testing and the Neyman-Pearson theory in the wider context of statistical methodology and theory.
Abstract: This is an account of the life of the author’s book Testing Statistical Hypotheses , its genesis, philosophy, reception and publishing history. There is also some discussion of the position of hypothesis testing and the Neyman-Pearson theory in the wider context of statistical methodology and theory.
TL;DR: In celebration of 50 years of biometry at the National Institutes of Health (NIH), a conference was held in January 1993 to acknowledge the contributions of those pioneers who laid the foundation in the 1940s and continued, through their seminal contributions, persuasiveness and perseverance, to foster the strong presence of biostatistics at NIH that exists today.
Abstract: In celebration of 50 years of biometry at the National Institutes of Health (NIH), a conference was held in January 1993 to acknowledge the contributions of those pioneers who laid the foundation in the 1940s and continued, through their seminal contributions, persuasiveness and perseverance, to foster the strong presence of biostatistics at NIH that exists today [5]. Biostatistics first appeared as a recognized discipline at the National Institutes of Health in the years 1946-1948. The Division of Statistical Methods in the United States Public Health Service was established with Harold Dorn as its first Head to support the research of the then new NIH. The degree of formal statistical training of his first recruits (Jerry Cornfield, Sam Greenhouse, Jack Lieberman, Nathan Mantel and Marvin Schneiderman) varied, but their experiencein the applications of statistics to problems of biology and medicine was minimal [12, 131. Within a few years, Sid Cutler, Max Halperin, Bill Haenszel, Harold Kahn, Sam Marcus, Felix Moore and others rounded out the starting team. Morton Kramer headed a statistics group at the separate National Institute of Mental Health.
TL;DR: Greenhouse as mentioned in this paper is a Fellow of the American Statistical Association, the Institute of Mathematical Statistics, the American Association for the Advancement of Science (AAAS), and an elected Fellow of The Royal Statistical Society (RSS).
Abstract: Samuel Greenhouse was born on January 13, 1918, in the Bronx (New York). He received a B.S. degree in mathematics from the City College of New York in 1938, an M.A. degree from George Washington University in 1954 and a Ph.D. in mathematical statistics from George Washington University in 1959. He is a Fellow of the American Statistical Association, the Institute of Mathematical Statistics, the American Association for the Advancement of Science (AAAS) and an elected Fellow of the Royal Statistical Society. He is also an elected member of the International Statistical Institute, a Fellow of the American College of Epidemiology and a Fellow of the Council of Epidemiology, American Heart Association. He is a past President of the Eastern North American Region of the Biometric Society and served on the Council of the International Biometric Society. He has served as President of the Washington Statistical Society, as Chairman of Section U (Statistics) of the AAAS and as a member of the AAAS Council Executive Committee. He has been an Associate Editor of the Journal of the American Statistical Association, and has served on the Board of Directors of the Society for Clinical Trials. His tenure at the National Institutes of Health included the years 1948-1974, where he began as a mathematical statistician at the National Cancer Institute. He served next as Chief of the Theoretical Statistics and Mathematics Section in the Biometry Branch of the National Institute of Mental Health (1954-1966), with an interlude as Visiting Professor of Statistics at Stanford University. He joined the National Institute of Child Health and Human Development (NICHD) in 1966 as the Chief of the Epidemiology and Biometry Branch. He wore two hats at NICHD, as Associate Director for Epidemiology and Biometry and as Acting Associate Director for Program Planning and Evaluation at the time of his retirement from NIH in 1974. Since leaving the NIH, he has been Professor of Statistics at George Washington University, serving as Head of the Department of Statistics 1976-1979 and 1986. During this time he was also Visiting Professor of Biostatistics at the Harvard School of Public Health. He is currently Associate Director for Research Development at the Biostatistics Center of George Washington University, and Professor Emeritus, George Washington University.
Abstract: Three papers from the early history of efficient parametric estimation are reprinted with commentary: (1) Fisher (1912), "On an absolute criterion for fitting frequency curves"; (2) Engledow and Yule (1914), "The determination of the best value of the coupling-ratio from a given set of data"; and (3) Fisher (1922), "The systematic location of genes by means of crossover observations."
TL;DR: Mantel was a Research Professor of Statistics at George Washington University and currently holds the title of Research Professor at American University as mentioned in this paper, where he has served on the editorial boards of Risk Analysis, Biometrics, Journal of the National Cancer Institute and Cancer Research.
Abstract: Nathan Mantel was born on February 16, 1919, in New York City. He received a B.S. degree in statistics from the City College of New York in 1939 and an M.A. degree from American University in 1956. He is a Fellow of the American Statistical Association and the Institute of Mathematical Statistics, has been elected Fellow of the Royal Statistical Society (RSS) and was recently made an Honorary Fellow of the RSS. He is also an elected member of the International Statistical Institute and a Fellow of the American Association for the Advancement of Science. He has been President of the Eastern North American Region of the Biometric Society and a member of the Council of the International Biometric Society. He has served on the editorial boards of Risk Analysis, Biometrics, Journal of the National Cancer Institute and Cancer Research. His tenure at the National Institutes of Health (NIH) included the years 1947-1974. This time was spent entirely as a mathematical statistician at the National Cancer Institute. While at NIH, he also held the position of Adjunct Professor of Biostatistics, Graduate School of Public Health, University of Pittsburgh. He was a recipient of the Superior Service Award, one of the highest civilian awards given by NIH. Since leaving NIH, he has been a Research Professor of Statistics at George Washington University and currently holds the title of Research Professor of Statistics at American University. Concurrently, from 1984 through 1990, he was Visiting Professor, Neuroepidemiology, at Temple University School of Medicine.
TL;DR: In this article, it is shown that the expected result of fitting the linear growth curve model to data that follow the alternative model is an ap- parent negative correlation between slope and intercept, and that the negative estimates of the correlation found in children's blood pressure data are an artifact of assuming a constant rate of change when the data actually follow the native model.
Abstract: Some analyses of longitudinal blood pressure data have fo- cussed on the question of whether a current value of blood pressure is predictive of subsequent rate of change. A positive correlation between blood pressure values at the beginning of a longitudinal study and rate of change over the course of the study has been found in studies of adults. Negative correlation, however, has been found in a study of children. These studies, either implicitly or explicitly, rely on linear growth curve models in which subjects' blood pressure observations are assumed to fol- low simple linear regression models with slopes and intercepts varying among subjects, but with the slopes constant over time. Our analysis of a longitudinal data set of 2,203 measurements of sys- tolic blood pressure from 216 children also provided a negative estimate of the correlation. However, smoothed plots of cross products of residu- als suggested that an alternative random effects model, in which rate of change of systolic blood pressure is not treated as constant over time, might better fit the data. It is possible that the negative estimates of the correlation found in children's blood pressure data are an artifact of assuming a constant rate of change when the data actually follow the al- ternative model. It is shown that the expected result of fitting the linear growth curve model to data that follow the alternative model is an ap- parent negative correlation between slope and intercept. In the data, the observed estimates of the parameters of the linear growth curve model are consistent with the observed estimates of the parameters of the al- ternative model.
TL;DR: The authors trace the emergence of Fisher's thinking on likelihood over a 10-year period and show that the typical way to think about statistical inference was in terms of the method of inverse probability and Bayes's theorem.
Abstract: When R. A. Fisher studied statistics as a student at Cambridge, the typical way to think about statistical inference was in terms of the method of inverse probability and Bayes's theorem. While others groped for alternatives with systematic structure and desirable alternatives, it remained for Fisher to invent the notion of likelihood and to explore its properties. These two papers trace the emergence of Fisher's thinking on likelihood over a 10-year period.
TL;DR: Morton Kramer served as Chief of the Biometrics Branch of the Na- tional Institute of Mental Health from 1949 through 1975, and in 1984 became Professor Emeritus at the National Institutes of Health.
Abstract: Morton Kramer was born on March 21, 1914, in Baltimore, Maryland. He received an A.B. degree from Johns Hopkins University in 1934 and a D.Sc. in 1939. He is a Fellow of the American Statistical Association, an Honorary Fellow of the American Psychiatric Associa- tion and an elected member of the Institute of Medicine of the National Academy of Sciences. He is the recipient of the Health-For-All Medal of the World Health Organization and the Distinguished Alumnus Award from Johns Hopkins University. During his tenure at NIH from 1949 through 1975, he served as Chief of the Biometrics Branch of the Na- tional Institute of Mental Health. He was awarded both the Department of Health, Education and Welfare Superior Service and Distinguished Service awards. Since leaving NIH, he was Professor of Biostatistics at the School of Hygiene and Public Health at Johns Hopkins University and in 1984 became Professor Emeritus. Ellenberg: When and why did you come to the National Institutes of Health (NIH) and what was your educational background? Kramer: I joined NIH in May of 1949. The Na- tional Institute of Mental Health (NIMH) was es- tablished in 1946 and Bob Felix, its first Director, brought me on board to be Chief of the Biometrics Branch.
TL;DR: Fred Ederer was a Senior Epidemiologist at the EMMES Corporation and Adjunct Professor in the Division of Biostatistics at the University of Minnesota and Lead Research Investigator, Clinical and Diagnostic Trials Section, Biometry Branch, Division of Cancer Prevention and Control, National Institutes of Health.
Abstract: Fred Ederer was born on March 5, 1926, in Vienna, Austria. He received a B.S. degree in mathematics and science from the City College of New York, an M.A. degree in statistics from American University and did further graduate work in biostatistics at Columbia and Stanford Universities. He is a Fellow of the American Statistical Association, of the American College of Epidemiology and of the American Heart Association's Council on Epidemiology. He has been on the editorial boards of the American Journal of Ophthalmology, Survey of Ophthalmology and the American Journal of Epidemiology. He has served on the Council on Epidemiology, the American Heart Association and the Regional Advisory Board of the Eastern North American Region of the Biometric Society, and he was on the Founding Board of Directors for both the Society for Clinical Trials and the American College of Epidemiology. His tenure at the National Institutes of Health (NIH) included the years 1957 through 1986. He began at the National Cancer Institute, moving next to the National Heart Institute and then spending the next half of his NIH career at the National Eye Institute (NEI). His first position at NEI was as Head of the Section on Clinical Trials, then Chief of the Office of Biometry and Epidemiology and, finally, Associate Director for Biometry and Epidemiology. He was awarded the Superior Service Award, one of the highest civilian awards given at NIH. Since leaving NIH, he has been Senior Epidemiologist at the EMMES Corporation and Adjunct Professor in the Division of Biostatistics at the University of Minnesota.
TL;DR: Harald Bergstrom was born on April 1, 1908, in Molltorp, situated in the middle of Sweden as mentioned in this paper, and studied at the University of Uppsala from 1932 to 1934.
Abstract: Harald Bergstrom was born on April 1, 1908, in Molltorp, situated in the middle of Sweden. A Master of Philosophy degree was awarded in 1931 at the University of Uppsala (in mathematics, physics and chemistry, extended to theoretical physics in 1932). He taught in secondary schools (gymnasiums) from 1932 to 1934, and then returned to Uppsala to pursue his research in mathematics toward a doctorate degree, which he received in 1938. He had a permanent lectureship in mathematics at the University of Uppsala from 1938 to 1945, and at a military college in 1945. In 1946, he was asked to hold a new established professorship in applied mathematics at the Chalmers University of Technology in Gothenburg, where he became full Professor in 1949. In 1960, the professorship in applied mathematics was expanded to two professorships--one in numerical analysis and one in mathematical statistics; he occupied the latter until his retirement in 1974.
TL;DR: Schneiderman was a statistician at the National Cancer Institute from 1948 through 1980 and later became Associate Director for Field Studies and Statistics at the University of California, Los Angeles as discussed by the authors.
Abstract: Marvin A. Schneiderman was born on December 25, 1918, in Brooklyn, New York. He received a B.S. degree in mathematics and statistics from the City College of New York in 1939, an M.S. degree in statistics from American University in 1953 and a Ph.D. in statistics from American University in 1961. Additional graduate training and research was done at Ohio State University, Harvard Graduate School of Business and the London School of Hygiene and Tropical Medicine. He is a Fellow of the American Statistical Association and of the American Association for the Advancement of Science. He is also an elected member of the International Statistical Institute, an elected Fellow of the Royal Statistical Society and a Founding Member of the American Society of Preventive Oncology. He has served as President of the Washington Statistical Society, as Chairman of the Committee on Presidents of Statistical Societies, as a member of the Board of Directors of the American Statistical Association and as a Council member of the International Biometric Society. He has been an editor on the editorial advisory boards of several journals, including Cancer Research, Statistics in Medicine, Blood, Journal of the National Cancer Institute and the American Journal of Industrial Medicine. He was at the National Cancer Institute from 1948 through 1980. He began as a consulting statistician, then was appointed Head of the Controlled Trials Group for Cancer Chemotherapy and later became Associate Director for Field Studies and Statistics. His last appointment at NIH was as NCI Associate Director for Science Policy. He was awarded two of the highest honors accorded civilian employees at the NIH, the Distinguished Service Award and the Superior Service Award. After leaving the National Institutes of Health, he spent a short time with a private consulting firm with strong environmental interests. He then served as a fellow at the Environmental Law Institute before joining the National Research Council/National Academy of Sciences Board on Environmental Studies and Toxicology. He officially retired in 1995. Marvin Schneiderman passed away on April 1, 1997.
TL;DR: John Christian Bailar III was a senior Scientist at NIH, a Senior Scientist at the Environmental Protection Agency and the Department of Health and Human Services, a Lecturer in Biostatistics at the Harvard School of Public Health, on staff at the Health Effects Institute and Professor and Chair of the Department at McGill University Faculty of Medicine.
Abstract: John Christian Bailar III was born on October 9, 1932, in Urbana, Illinois. He received his B.A. degree from the University of Colorado in 1953, an M.D. from Yale University in 1955 and a Ph.D. in statistics from American University in 1973. He is a Fellow of the American Statistical Association, an elected member of the International Statistical Institute, a Fellow of the American College of Epidemiology, a Fellow of the American Association for the Advancement of Science, an elected member of the Collegium Ramazzini and a MacArthur Fellow (1990-1995). He was Editor-in-Chief of the Journal of the National Cancer Institute and has been on the Editorial Board of Cancer Research. He has served as Statistical Consultant for the New England Journal of Medicine and is currently a member of its Editorial Board. He has served as Chair of the Biometrics Section of the American Statistical Association, was Founding Chair of the Boston Chapter of the Society for Risk Analysis and was President of the Council of Biology Editors. His tenure at NIH included the years 1956-1970 and 1972-1980 on staff at the National Cancer Institute, with a stint at the Veterans Administration from 1970-1972. He began as a Field Investigator in the Biometry Branch, was appointed Head of the Demography Section and then Director of the Third National Cancer Survey. His last appointment at NCI was Deputy Associate Director for Cancer Control. He was awarded the Commendation Medal from the United States Public Health Service for his work on breast cancer screening. Since leaving NIH, he was a Senior Scientist at the Environmental Protection Agency and the Department of Health and Human Services, a Lecturer in Biostatistics at the Harvard School of Public Health, on staff at the Health Effects Institute and Professor and Chair of the Department of Epidemiology and Biostatistics at McGill University Faculty of Medicine. Since 1995, he has been Professor and Chair of the Department of Health Studies at the University of Chicago.
TL;DR: Haenszel was an elected Fellow of the American Statistical Association, the American Public Health Association and the American Association For the Advancement of Science as discussed by the authors, and has been awarded a Doctor Honoris Causa en Salud Publica from the Universidad del Valle in Colombia.
Abstract: William M. Haenszel was born on June 19, 1910, in Rochester, New York. He received a B.A. degree in 1931 and an M.A. degree in 1932, both from the University of Buffalo. He is an elected Fellow of the American Statistical Association, the American Public Health Association and the American Association For the Advancement of Science. He has been awarded a Doctor Honoris Causa en Salud Publica from the Universidad del Valle in Colombia. He has held positions as Secretary of the Statistics Section and member of the governing Council of the American Public Health Association, Chair of the Biometrics Section of the American Statistical Association and member of the Regional Advisory Board of the Eastern North Atlantic Region of the International Biometric Society. During his tenure at the National Institutes of Health (NIH) from 1952 through 1976, he served as Head of the Biometric Section and the Chief of the Biometry Branch at the National Cancer Institute. Since leaving the National Institutes of Health he was on staff at the Illinois Cancer Council, Professor of Epidemiology at the University of Illinois (he is currently Professor Emeritus) and a consultant to the World Health Organization.
TL;DR: Sujit Kumar Mitra as discussed by the authors has made pioneering contributions in many areas of statistics and mathematics, including survey sampling, linear models, design of experiments, goodness-of-fit tests and linear algebra.
Abstract: Sujit Kumar Mitra was born on January 23, 1932, in Calcutta, India. He earned his B.Sc. degree in statistics from Presidency College, Calcutta, in 1949, an M.Sc. degree in statistics from Calcutta University in 1951 and a Ph.D. degree in statistics from the University of North Carolina at Chapel Hill in 1956, under the guidance of Professor S. N. Roy. He has made pioneering contributions in many areas of statistics and mathematics--including survey sampling, linear models, design of experiments, goodness-of-fit tests and linear algebra. He has been particularly credited for path-breaking contributions in the area of generalized inverses of matrices, culminating in a jointly authored landmark book with Professor C. R. Rao (published in 1971). He was Professor at the Indian Statistical Institute (ISI), both in Calcutta and Delhi, for almost 40 years. He has held visiting positions at Indiana University, Purdue University, University of Texas at Dallas, and Keio University, Japan. He retired from ISI in January 1992 and is currently Professor Emeritus. Unfortunately, he contracted Parkinson's disease in 1978. Despite an uphill battle against constant physical discomfort, he has continued as a leading contributor in many directions of mathematics and statistics. He is well known for his zest in attacking and solving some of the most difficult problems in linear algebra. He has earned many awards, honors and titles, including Fellow of the Indian National Science Academy and of the Indian Academy of Science. He was also elected President of the Statistics Section of the Indian Science Congress in 1988. He has been associated with many journals, including Sankhyā, and has edited or coedited several special volumes.
TL;DR: Tavia Gordon was heavily involved in the development of the design and analysis of the first long-term, large-scale community-based follow-up study in the United States, the Framingham Study.
Abstract: Tavia Gordon was born on December 14, 1917, in Chicago, Illinois. He received a B.A. degree in anthropology from the University of California in 1938. He did graduate work in anthropology at the University of Chicago in 1938-1939, in mathematics at the University of Southern California in 1947-1948, and in mathematical statistics at the University of California, Berkeley, in 1948-1950. He is a Fellow of the American Statistical Association and the Council on Epidemiology, American Heart Association. His tenure at NIH included the years 1954-1960 and 1966-1977, beginning as an Analytical Statistician with the Biometrics Research Section of the National Heart Institute. He spent the next two years at the Biometry Branch at the National Cancer Institute. His last 10 years at NIH were spent at the National Heart, Lung and Blood Institute. During this period he was heavily involved in the development of the design and analysis of the first long-term, large-scale community-based follow-up study in the United States, the Framingham Study. He was awarded the NIH Director's Award in 1977. Since leaving the National Institutes of Health, he has been a consulting statistician, a senior scientist for General Electric Corporation and, since 1981, a Research Professor at George Washington University Biostatistics Center.