Tze Leung Lai
Member, Wu Tsai Neurosciences Institute
Bio
Tze Lai was the Ray Lyman Wilbur Professor of Statistics and, by courtesy, of Biomedical Data Science and of the Institute of Computational and Mathematical Engineering (ICME) at Stanford University. He served as Co-director of the Center for Innovative Study Design (CISD) at the Stanford School of Medicine. Professor Lai supervised more than 73 Ph.D. students and published over 300 papers in topics covering clinical trial design and analysis, population pharmacokinetics and pharmacodynamics, survival analysis, longitudinal data analysis, multiple endpoints, biomarkers, adaptive methods, sequential analysis and time series.
Administrative Appointments
-
Core Member, Center for Precision Mental Health and Wellness, Stanford University (2018 - Present)
-
Faculty Fellow, Center for Innovations in Global Health, Stanford University (2018 - Present)
-
Ray Lyman Wilbur Professor, Stanford University (2017 - Present)
-
Professor, by courtesy, of Biomedical Data Science (formerly Health Research and Policy), Stanford School of Medicine (2015 - Present)
-
Co-chair, Program in Mathematical and Computational Finance, Stanford University (2014 - Present)
-
Director, Financial and Risk Modeling Institute,, Stanford University (2012 - Present)
-
International Advisory Committee Member, Mathematical Sciences Center, Tsinghua University (2012 - Present)
-
Advisory Committee Member, Institute of Mathematical Research, University of Hong Kong (2011 - Present)
-
Chair, Pao-Lu Hsu Distinguished Lecture Series in Statistics and Probability, Center for Mathematical Sciences, Tsinghua University (2011 - Present)
-
International Advisory Committee Member, Center for Statistical Sciences, Peking University (2010 - Present)
-
Co-director, Center for Innovative Study Design, Stanford School of Medicine (2009 - Present)
-
Professor, by courtesy, of the Institute of Computational and Mathematical Engineering, Stanford School of Engineering (2009 - Present)
-
Professor, by courtesy, of Health Research and Policy, Stanford School of Medicine (2007 - 2015)
-
Steering Committee, Methods of Analysis Program in the Social Sciences, Stanford University (2005 - Present)
-
Co-director, Biostatistics Core, Stanford Cancer Center (2005 - 2016)
-
Director, Interdisciplinary Program in Financial Mathematics, Stanford University (2005 - 2014)
-
Chair, School of Statistics, Chinese Academy of Sciences (2003 - Present)
Professional Education
-
B.A. with First Class Honors, University of Hong Kong, Mathematics (1967)
-
M.A., Columbia University, Mathematical Statistics (1970)
-
Ph.D., Columbia University, Mathematical Statistics (1971)
Current Research and Scholarly Interests
Tze Lai was widely recognized as a prolific leader in the field of sequential statistical analysis. Among his principal achievements was the development of a comprehensive theory of sequential tests of composite hypotheses, unifying previous approaches and providing far-reaching extensions to cope with the practical complexities that arise in the applications to group sequential clinical trials. In particular, this theory paved the way for his ground-breaking work with Shih on flexible and nearly-optimal group sequential tests that can "self-tune" to the unknown parameters during the course of the trial, under pre-specified constraints on the maximum sample size and significance level.
Other major breakthroughs include (a) accurate confidence intervals following sequential tests by using an innovative resampling approach, (b) a definitive solution to the long-standing "multi-armed bandit problem", and (c) the development of statistically and computationally efficient sequential change-point detection procedures in multivariate time series and stochastic systems, for applications to industrial quality control, fault detection in engineering systems and segmentation in computational biology.
Besides sequential analysis, Lai also made ground-breaking contributions to (i) stochastic approximation and recursive estimation, (ii) adaptive control of linear stochastic systems and Markov decision processes, (iii) saddlepoint approximations and boundary-crossing probabilities in Markov random walks and random fields, and (iv) survival analysis, in particular, rand- and M-estimators in regression models when the response variable is subject to censoring and truncation, and interim analysis of clinical trials with failure-time endpoints.
2022-23 Courses
-
Independent Studies (7)
- Independent Study
STATS 199 (Aut) - Independent Study
STATS 299 (Aut, Win, Spr) - Industrial Research for Statisticians
STATS 298 (Aut, Win) - Industrial Research for Statisticians
STATS 398 (Aut, Win, Spr, Sum) - Master's Research
CME 291 (Aut) - Ph.D. Research
CME 400 (Aut, Win, Spr, Sum) - Research
STATS 399 (Aut, Win, Spr, Sum)
- Independent Study
-
Prior Year Courses
2021-22 Courses
- Neuroscience-inspired Optimization in Artificial Intelligence, with Multifaceted Applications
BIOMEDIN 346, STATS 246 (Win) - Risk Analytics and Management in Finance and Insurance
CME 243, STATS 243 (Win) - Statistical Methods in Finance
STATS 240 (Aut)
- Neuroscience-inspired Optimization in Artificial Intelligence, with Multifaceted Applications
All Publications
-
Power of an Adaptive Trial Design for Endovascular Stroke Studies Simulations Using IMS (Interventional Management of Stroke) III Data
STROKE
2016; 47 (12): 2931-2937
Abstract
Adaptive trial designs that allow enrichment of the study population through subgroup selection can increase the chance of a positive trial when there is a differential treatment effect among patient subgroups. The goal of this study is to illustrate the potential benefit of adaptive subgroup selection in endovascular stroke studies.We simulated the performance of a trial design with adaptive subgroup selection and compared it with that of a traditional design. Outcome data were based on 90-day modified Rankin Scale scores, observed in IMS III (Interventional Management of Stroke III), among patients with a vessel occlusion on baseline computed tomographic angiography (n=382). Patients were categorized based on 2 methods: (1) according to location of the arterial occlusive lesion and onset-to-randomization time and (2) according to onset-to-randomization time alone. The power to demonstrate a treatment benefit was based on 10 000 trial simulations for each design.The treatment effect was relatively homogeneous across categories when patients were categorized based on arterial occlusive lesion and time. Consequently, the adaptive design had similar power (47%) compared with the fixed trial design (45%). There was a differential treatment effect when patients were categorized based on time alone, resulting in greater power with the adaptive design (82%) than with the fixed design (57%).These simulations, based on real-world patient data, indicate that adaptive subgroup selection has merit in endovascular stroke trials as it substantially increases power when the treatment effect differs among subgroups in a predicted pattern.
View details for DOI 10.1161/STROKEAHA.116.015436
View details for Web of Science ID 000389424200022
View details for PubMedID 27895297
View details for PubMedCentralID PMC5134921
-
MULTIVARIATE STOCHASTIC REGRESSION IN TIME SERIES MODELING
STATISTICA SINICA
2016; 26 (4): 1411-1426
View details for DOI 10.5705/ss.2014.211t
View details for Web of Science ID 000387199200005
-
Discussion on "Sequential detection/isolation of abrupt changes" by Igor V. Nikiforov
SEQUENTIAL ANALYSIS-DESIGN METHODS AND APPLICATIONS
2016; 35 (3): 305-310
View details for DOI 10.1080/07474946.2016.1206372
View details for Web of Science ID 000384212400004
-
Adaptive design of confirmatory trials: Advances and challenges
CONTEMPORARY CLINICAL TRIALS
2015; 45: 93-102
Abstract
The past decade witnessed major developments in innovative designs of confirmatory clinical trials, and adaptive designs represent the most active area of these developments. We give an overview of the developments and associated statistical methods in several classes of adaptive designs of confirmatory trials. We also discuss their statistical difficulties and implementation challenges, and show how these problems are connected to other branches of mainstream Statistics, which we then apply to resolve the difficulties and bypass the bottlenecks in the development of adaptive designs for the next decade.
View details for DOI 10.1016/j.cct.2015.06.007
View details for Web of Science ID 000367754700014
View details for PubMedCentralID PMC4639447
-
Innovative designs of point-of-care comparative effectiveness trials
CONTEMPORARY CLINICAL TRIALS
2015; 45: 61-68
Abstract
One of the provisions of the health care reform legislation in 2010 was for funding pragmatic clinical trials or large observational studies for comparing the effectiveness of different approved medical treatments, involving broadly representative patient populations. After reviewing pragmatic clinical trials and the issues and challenges that have made them just a small fraction of comparative effectiveness research (CER), we focus on a recent development that uses point-of-care (POC) clinical trials to address the issue of "knowledge-action gap" in pragmatic CER trials. We give illustrative examples of POC-CER trials and describe a trial that we are currently planning to compare the effectiveness of newly approved oral anticoagulants. We also develop novel stage-wise designs of information-rich POC-CER trials under competitive budget constraints, by using recent advances in adaptive designs and other statistical methodologies.
View details for DOI 10.1016/j.cct.2015.06.014
View details for Web of Science ID 000367754700010
View details for PubMedCentralID PMC4639459
-
ABO Mismatch Is Associated with Increased Nonrelapse Mortality after Allogeneic Hematopoietic Cell Transplantation
BIOLOGY OF BLOOD AND MARROW TRANSPLANTATION
2015; 21 (4): 746-754
Abstract
We evaluated ABO associated outcomes in 1737 patients who underwent allogeneic hematopoietic cell transplantation (allo-HCT) at Stanford University between January 1986 and July 2011. Grafts were 61% ABO matched, 18% major mismatched (MM), 17% minor MM, and 4% bidirectional MM. Median follow-up was 6 years. In multivariate analysis, overall survival (OS) was inferior in minor MM hematopoietic cell transplantations (median 2.1 versus 6.3 years; hazard ratio [HR], 1.56; 95% confidence interval [CI], 1.19 to 2.05; P = .001) in comparison with ABO-matched grafts. ABO minor MM was associated with an increase in early nonrelapse mortality (NRM) (18% versus 13%; HR, 1.48; 95% CI, 1.06 to 2.06; P = .02). In an independent Center for International Blood and Marrow Transplant Research (CIBMTR) analysis of 435 lymphoma patients receiving mobilized peripheral blood grafts, impairment of OS (HR, 1.55; 95% CI, 1.07 to 2.25; P = .021) and increased NRM (HR, 1.72; 95% CI, 1.11 to 2.68; P = .03) were observed in recipients of ABO minor-MM grafts. A second independent analysis of a CIBMTR data set including 5179 patients with acute myeloid leukemia and myelodysplastic syndrome identified a nonsignificant trend toward decreased OS in recipients of ABO minor-MM grafts and also found ABO major MM to be significantly associated with decreased OS (HR, 1.19; 95% CI, 1.08 to 1.31; P < .001) and increased NRM (HR, 1.23; 95% CI, 1.08 to 1.4; P = .002). ABO minor and major MM are risk factors for worse transplantation outcomes, although the associated hazards may not be uniform across different transplantation populations. Further study is warranted to determine which patient populations are at greatest risk, and whether this risk can be modified by anti-B cell therapy or other peri-transplantation treatments.
View details for DOI 10.1016/j.bbmt.2014.12.036
View details for PubMedID 25572032
-
ASYMPTOTICALLY EFFICIENT PARAMETER ESTIMATION IN HIDDEN MARKOV SPATIO-TEMPORAL RANDOM FIELDS
STATISTICA SINICA
2015; 25 (1): 403-421
View details for DOI 10.5705/ss.2013.281w
View details for Web of Science ID 000348969700024
-
DYNAMIC EMPIRICAL BAYES MODELS AND THEIR APPLICATIONS TO LONGITUDINAL DATA ANALYSIS AND PREDICTION
STATISTICA SINICA
2014; 24 (4): 1505-1528
View details for DOI 10.5705/ss.2012.048
View details for Web of Science ID 000345201200002
-
Discussion on "Sequential Estimation for Time Series Models" by T. N. Sriram and Ross Iaci
SEQUENTIAL ANALYSIS-DESIGN METHODS AND APPLICATIONS
2014; 33 (2): 169-173
View details for DOI 10.1080/07474946.2014.896688
View details for Web of Science ID 000335952800005
-
A GENERAL THEORY OF PARTICLE FILTERS IN HIDDEN MARKOV MODELS AND SOME APPLICATIONS
ANNALS OF STATISTICS
2013; 41 (6): 2877-2904
View details for DOI 10.1214/13-AOS1172
View details for Web of Science ID 000330204900007
-
Group sequential designs for developing and testing biomarker-guided personalized therapies in comparative effectiveness research.
Contemporary clinical trials
2013; 36 (2): 651-663
Abstract
Biomarker-guided personalized therapies offer great promise to improve drug development and improve patient care, but also pose difficult challenges in designing clinical trials for the development and validation of these therapies. We first give a review of the existing approaches, briefly for clinical trials in new drug development and in more detail for comparative effectiveness trials involving approved treatments. We then introduce new group sequential designs to develop and test personalized treatment strategies involving approved treatments.
View details for DOI 10.1016/j.cct.2013.08.007
View details for PubMedID 23994669
-
Group sequential designs for developing and testing biomarker-guided personalized therapies in comparative effectiveness research.
Contemporary clinical trials
2013; 36 (2): 651-663
Abstract
Biomarker-guided personalized therapies offer great promise to improve drug development and improve patient care, but also pose difficult challenges in designing clinical trials for the development and validation of these therapies. We first give a review of the existing approaches, briefly for clinical trials in new drug development and in more detail for comparative effectiveness trials involving approved treatments. We then introduce new group sequential designs to develop and test personalized treatment strategies involving approved treatments.
View details for DOI 10.1016/j.cct.2013.08.007
View details for PubMedID 23994669
-
STOCHASTIC CHANGE-POINT ARX-GARCH MODELS AND THEIR APPLICATIONS TO ECONOMETRIC TIME SERIES
STATISTICA SINICA
2013; 23 (4): 1573-1594
View details for DOI 10.5705/ss.2012.224s
View details for Web of Science ID 000339125900009
-
RARE-EVENT SIMULATION OF HEAVY-TAILED RANDOM WALKS BY SEQUENTIAL IMPORTANCE SAMPLING AND RESAMPLING
ADVANCES IN APPLIED PROBABILITY
2012; 44 (4): 1173-1196
View details for Web of Science ID 000313538600011
-
Sequential design of phase II-III cancer trials
STATISTICS IN MEDICINE
2012; 31 (18): 1944-1960
Abstract
Although traditional phase II cancer trials are usually single arm, with tumor response as endpoint, and phase III trials are randomized and incorporate interim analyses with progression-free survival or other failure time as endpoint, this paper proposes a new approach that seamlessly expands a randomized phase II study of response rate into a randomized phase III study of time to failure. This approach is based on advances in group sequential designs and joint modeling of the response rate and time to event. The joint modeling is reflected in the primary and secondary objectives of the trial, and the sequential design allows the trial to adapt to increase in information on response and survival patterns during the course of the trial and to stop early either for conclusive evidence on efficacy of the experimental treatment or for the futility in continuing the trial to demonstrate it, on the basis of the data collected so far.
View details for DOI 10.1002/sim.5346
View details for Web of Science ID 000306471100004
View details for PubMedID 22422502
-
Clinical trial designs for testing biomarker-based personalized therapies
CLINICAL TRIALS
2012; 9 (2): 141-154
Abstract
Advances in molecular therapeutics in the past decade have opened up new possibilities for treating cancer patients with personalized therapies, using biomarkers to determine which treatments are most likely to benefit them, but there are difficulties and unresolved issues in the development and validation of biomarker-based personalized therapies. We develop a new clinical trial design to address some of these issues. The goal is to capture the strengths of the frequentist and Bayesian approaches to address this problem in the recent literature and to circumvent their limitations.We use generalized likelihood ratio tests of the intersection null and enriched strategy null hypotheses to derive a novel clinical trial design for the problem of advancing promising biomarker-guided strategies toward eventual validation. We also investigate the usefulness of adaptive randomization (AR) and futility stopping proposed in the recent literature.Simulation studies demonstrate the advantages of testing both the narrowly focused enriched strategy null hypothesis related to validating a proposed strategy and the intersection null hypothesis that can accommodate to a potentially successful strategy. AR and early termination of ineffective treatments offer increased probability of receiving the preferred treatment and better response rates for patients in the trial, at the expense of more complicated inference under small-to-moderate total sample sizes and some reduction in power.The binary response used in the development phase may not be a reliable indicator of treatment benefit on long-term clinical outcomes. In the proposed design, the biomarker-guided strategy (BGS) is not compared to 'standard of care', such as physician's choice that may be informed by patient characteristics. Therefore, a positive result does not imply superiority of the BGS to 'standard of care'. The proposed design and tests are valid asymptotically. Simulations are used to examine small-to-moderate sample properties.Innovative clinical trial designs are needed to address the difficulties and issues in the development and validation of biomarker-based personalized therapies. The article shows the advantages of using likelihood inference and interim analysis to meet the challenges in the sample size needed and in the constantly evolving biomarker landscape and genomic and proteomic technologies.
View details for DOI 10.1177/1740774512437252
View details for Web of Science ID 000302636500001
View details for PubMedID 22397801
-
Adaptive Trial Designs
ANNUAL REVIEW OF PHARMACOLOGY AND TOXICOLOGY, VOL 52
2012; 52: 101-110
Abstract
We review adaptive designs for clinical trials, giving special attention to the control of the Type I error in late-phase confirmatory trials, when the trial planner wishes to adjust the final sample size of the study in response to an unblinded analysis of interim estimates of treatment effects. We point out that there is considerable inefficiency in using the adaptive designs that employ conditional power calculations to reestimate the sample size and that maintain the Type I error by using certain weighted test statistics. Although these adaptive designs have little advantage over familiar group-sequential designs, our review also describes recent developments in adaptive designs that are both flexible and efficient. We also discuss the use of Bayesian designs, when the context of use demands control over operating characteristics (Type I and II errors) and correction of the bias of estimated treatment effects.
View details for DOI 10.1146/annurev-pharmtox-010611-134504
View details for Web of Science ID 000301839600006
View details for PubMedID 21838549
-
Adaptation in clinical development plans and adaptive clinical trial designs
STATISTICS AND ITS INTERFACE
2012; 5 (4): 431-442
View details for Web of Science ID 000312365700007
-
Sequential Importance Sampling and Resampling for Dynamic Portfolio Credit Risk
OPERATIONS RESEARCH
2012; 60 (1): 78-91
View details for DOI 10.1287/opre.1110.1008
View details for Web of Science ID 000302113900007
-
Futility stopping in clinical trials
STATISTICS AND ITS INTERFACE
2012; 5 (4): 415-423
View details for Web of Science ID 000312365700005
-
Efficient Adaptive Randomization and Stopping Rules in Multi-arm Clinical Trials for Testing a New Treatment
3rd International Workshop on Sequential Methodologies (IWSM)
TAYLOR & FRANCIS INC. 2012: 441–57
View details for DOI 10.1080/07474946.2012.719433
View details for Web of Science ID 000323818500002
-
A SEQUENTIAL MONTE CARLO APPROACH TO COMPUTING TAIL PROBABILITIES IN STOCHASTIC MODELS
ANNALS OF APPLIED PROBABILITY
2011; 21 (6): 2315-2342
View details for DOI 10.1214/10-AAP758
View details for Web of Science ID 000298249900009
-
EVALUATING PROBABILITY FORECASTS
ANNALS OF STATISTICS
2011; 39 (5): 2356-2382
View details for DOI 10.1214/11-AOS902
View details for Web of Science ID 000299186500007
-
A STEPWISE REGRESSION METHOD AND CONSISTENT MODEL SELECTION FOR HIGH-DIMENSIONAL SPARSE LINEAR MODELS
STATISTICA SINICA
2011; 21 (4): 1473-1513
View details for DOI 10.5705/ss.2010.081
View details for Web of Science ID 000296212600001
-
MEAN-VARIANCE PORTFOLIO OPTIMIZATION WHEN MEANS AND COVARIANCES ARE UNKNOWN
ANNALS OF APPLIED STATISTICS
2011; 5 (2A): 798-823
View details for DOI 10.1214/10-AOAS422
View details for Web of Science ID 000295453300010
-
Incorporating Individual and Collective Ethics into Phase I Cancer Trial Designs
BIOMETRICS
2011; 67 (2): 596-603
Abstract
A general framework is proposed for Bayesian model based designs of Phase I cancer trials, in which a general criterion for coherence (Cheung, 2005, Biometrika 92, 863-873) of a design is also developed. This framework can incorporate both "individual" and "collective" ethics into the design of the trial. We propose a new design that minimizes a risk function composed of two terms, with one representing the individual risk of the current dose and the other representing the collective risk. The performance of this design, which is measured in terms of the accuracy of the estimated target dose at the end of the trial, the toxicity and overdose rates, and certain loss functions reflecting the individual and collective ethics, is studied and compared with existing Bayesian model based designs and is shown to have better performance than existing designs.
View details for DOI 10.1111/j.1541-0420.2010.01471.x
View details for Web of Science ID 000292504000027
View details for PubMedID 20731643
-
A SIMPLE BAYESIAN APPROACH TO MULTIPLE CHANGE-POINTS
STATISTICA SINICA
2011; 21 (2): 539-569
View details for Web of Science ID 000290459900004
-
CRAMER TYPE MODERATE DEVIATIONS FOR STUDENTIZED U-STATISTICS
ESAIM-PROBABILITY AND STATISTICS
2011; 15: 168-179
View details for DOI 10.1051/ps/2009014
View details for Web of Science ID 000300777300008
-
Discussion on "Two-Stage Procedures for High-Dimensional Data" by Makoto Aoshima and Kazuyoshi Yata
SEQUENTIAL ANALYSIS-DESIGN METHODS AND APPLICATIONS
2011; 30 (4): 404-411
View details for DOI 10.1080/07474946.2011.619092
View details for Web of Science ID 000208760700004
-
Sequential generalized likelihood ratio tests for vaccine safety evaluation
STATISTICS IN MEDICINE
2010; 29 (26): 2698-2708
Abstract
The evaluation of vaccine safety involves pre-clinical animal studies, pre-licensure randomized clinical trials, and post-licensure safety studies. Sequential design and analysis are of particular interest because they allow early termination of the trial or quick detection that the vaccine exceeds a prescribed bound on the adverse event rate. After a review of the recent developments in this area, we propose a new class of sequential generalized likelihood ratio tests for evaluating adverse event rates in two-armed pre-licensure clinical trials and single-armed post-licensure studies. The proposed approach is illustrated using data from the Rotavirus Efficacy and Safety Trial. Simulation studies of the performance of the proposed approach and other methods are also given.
View details for DOI 10.1002/sim.4036
View details for Web of Science ID 000284023800004
View details for PubMedID 20799244
-
Approximate Dynamic Programming and Its Applications to the Design of Phase I Cancer Trials
STATISTICAL SCIENCE
2010; 25 (2): 245-257
View details for DOI 10.1214/10-STS317
View details for Web of Science ID 000285322300007
-
Multistage Tests of Multiple Hypotheses
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS
2010; 39 (8-9): 1597-1607
View details for DOI 10.1080/03610920802592852
View details for Web of Science ID 000277463700022
-
Theory and applications of multivariate self-normalized processes
STOCHASTIC PROCESSES AND THEIR APPLICATIONS
2009; 119 (12): 4210-4227
View details for DOI 10.1016/j.spa.2009.10.003
View details for Web of Science ID 000272328000010
-
Tests and confidence intervals for secondary endpoints in sequential clinical trials
BIOMETRIKA
2009; 96 (4): 903-915
View details for DOI 10.1093/biomet/asp063
View details for Web of Science ID 000272179100011
-
Option hedging theory under transaction costs
JOURNAL OF ECONOMIC DYNAMICS & CONTROL
2009; 33 (12): 1945-1961
View details for DOI 10.1016/j.jedc.2009.04.007
View details for Web of Science ID 000272262500001
-
A Bayesian Approach to Sequential Surveillance in Exponential Families
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS
2009; 38 (16-17): 2958-2968
View details for DOI 10.1080/03610920902947253
View details for Web of Science ID 000269147300025
-
Modern sequential analysis and its applications to computerized adaptive testing
PSYCHOMETRIKA
2008; 73 (3): 473-486
View details for DOI 10.1007/s11336-007-9053-9
View details for Web of Science ID 000259578700008
-
Statistical inference in dynamic panel data models
JOURNAL OF STATISTICAL PLANNING AND INFERENCE
2008; 138 (9): 2763-2776
View details for DOI 10.1016/j.jspi.2008.03.011
View details for Web of Science ID 000256602600015
-
Stochastic segmentation models for array-based comparative genomic hybridization data analysis
BIOSTATISTICS
2008; 9 (2): 290-307
Abstract
Array-based comparative genomic hybridization (array-CGH) is a high throughput, high resolution technique for studying the genetics of cancer. Analysis of array-CGH data typically involves estimation of the underlying chromosome copy numbers from the log fluorescence ratios and segmenting the chromosome into regions with the same copy number at each location. We propose for the analysis of array-CGH data, a new stochastic segmentation model and an associated estimation procedure that has attractive statistical and computational properties. An important benefit of this Bayesian segmentation model is that it yields explicit formulas for posterior means, which can be used to estimate the signal directly without performing segmentation. Other quantities relating to the posterior distribution that are useful for providing confidence assessments of any given segmentation can also be estimated by using our method. We propose an approximation method whose computation time is linear in sequence length which makes our method practically applicable to the new higher density arrays. Simulation studies and applications to real array-CGH data illustrate the advantages of the proposed approach.
View details for DOI 10.1093/biostatistics/kxm031
View details for Web of Science ID 000254293400007
View details for PubMedID 17855472
-
Statistical models for the Basel II internal ratings-based approach to measuring credit risk of retail products
STATISTICS AND ITS INTERFACE
2008; 1 (2): 229-241
View details for Web of Science ID 000207654800003
-
A Hidden Markov Filtering Approach to Multiple Change-point Models
47th IEEE Conference on Decision and Control
IEEE. 2008: 1914–1919
View details for Web of Science ID 000307311602007
-
Corrected random walk approximations to free boundary problems in optimal stopping
ADVANCES IN APPLIED PROBABILITY
2007; 39 (3): 753-775
View details for Web of Science ID 000250440800008
-
Efficient importance sampling for Monte Carlo evaluation of exceedance probabilities
ANNALS OF APPLIED PROBABILITY
2007; 17 (2): 440-473
View details for DOI 10.1214/105051606000000664
View details for Web of Science ID 000245843200002
-
A combined superiority and non-inferiority approach to multiple endpoints in clinical trials
STATISTICS IN MEDICINE
2007; 26 (6): 1193-1207
Abstract
Treatment comparisons in clinical trials often involve multiple endpoints. By making use of bootstrap tests, we develop a new non-parametric approach to multiple-endpoint testing that can be used to demonstrate non-inferiority of a new treatment for all endpoints and superiority for some endpoint when it is compared to an active control. It is shown that this approach does not incur a large multiplicity cost in sample size to achieve reasonable power and that it can incorporate complex dependencies in the multivariate distributions of all outcome variables for the two treatments via bootstrap resampling.
View details for DOI 10.1002/sim.2611
View details for Web of Science ID 000244903400002
View details for PubMedID 16791905
-
Nonparametric functionals of spectral distributions and their applications to time series analysis
JOURNAL OF STATISTICAL PLANNING AND INFERENCE
2007; 137 (3): 1066-1075
View details for DOI 10.1016/j.jspi.2006.06.025
View details for Web of Science ID 000242706300032
-
Marginal regression analysis of longitudinal data with time-dependent covariates: a generalized method-of-moments approach
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY
2007; 69: 79-99
View details for Web of Science ID 000243479600005
-
Identification and adaptive control of change-point ARX models via Rao-Blackwellized particle filters
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
2007; 52 (1): 67-72
View details for DOI 10.1109/TAC.2006.887913
View details for Web of Science ID 000243619900006
-
Confidence intervals for survival quantiles in the Cox regression model
LIFETIME DATA ANALYSIS
2006; 12 (4): 407-419
Abstract
Median survival times and their associated confidence intervals are often used to summarize the survival outcome of a group of patients in clinical trials with failure-time endpoints. Although there is an extensive literature on this topic for the case in which the patients come from a homogeneous population, few papers have dealt with the case in which covariates are present as in the proportional hazards model. In this paper we propose a new approach to this problem and demonstrate its advantages over existing methods, not only for the proportional hazards model but also for the widely studied cases where covariates are absent and where there is no censoring. As an illustration, we apply it to the Stanford Heart Transplant data. Asymptotic theory and simulation studies show that the proposed method indeed yields confidence intervals and bands with accurate coverage errors.
View details for DOI 10.1007/s10985-006-9024-y
View details for Web of Science ID 000242998200002
View details for PubMedID 17053975
-
Confidence intervals in group sequential trials with random group sizes and applications to survival analysis
BIOMETRIKA
2006; 93 (3): 641-654
View details for Web of Science ID 000241271700011
-
Efficient recursive estimation and adaptive control in stochastic regression and ARMAX models
STATISTICA SINICA
2006; 16 (3): 741-772
View details for Web of Science ID 000240123900006
-
Modified Haybittle-Peto group sequential designs for testing superiority and non-inferiority hypotheses in clinical trials
STATISTICS IN MEDICINE
2006; 25 (7): 1149-1167
Abstract
In designing an active controlled clinical trial, one sometimes has to choose between a superiority objective (to demonstrate that a new treatment is more effective than an active control therapy) and a non-inferiority objective (to demonstrate that it is no worse than the active control within some pre-specified non-inferiority margin). It is often difficult to decide which study objective should be undertaken at the planning stage when one does not have actual data on the comparative advantage of the new treatment. By making use of recent advances in the theory of efficient group sequential tests, we show how this difficulty can be resolved by a flexible group sequential design that can adaptively choose between the superiority and non-inferiority objectives during interim analyses. While maintaining the type I error probability at a pre-specified level, the proposed test is shown to have power advantage and/or sample size saving over fixed sample size tests for either only superiority or non-inferiority, and over other group sequential designs in the literature.
View details for DOI 10.1002/sim.2357
View details for Web of Science ID 000236528500005
View details for PubMedID 16189814
-
Flexible modeling via a hybrid estimation scheme in generalized mixed models for longitudinal data
BIOMETRICS
2006; 62 (1): 159-167
Abstract
To circumvent the computational complexity of likelihood inference in generalized mixed models that assume linear or more general additive regression models of covariate effects, Laplace's approximations to multiple integrals in the likelihood have been commonly used without addressing the issue of adequacy of the approximations for individuals with sparse observations. In this article, we propose a hybrid estimation scheme to address this issue. The likelihoods for subjects with sparse observations use Monte Carlo approximations involving importance sampling, while Laplace's approximation is used for the likelihoods of other subjects that satisfy a certain diagnostic check on the adequacy of Laplace's approximation. Because of its computational tractability, the proposed approach allows flexible modeling of covariate effects by using regression splines and model selection procedures for knot and variable selection. Its computational and statistical advantages are illustrated by simulation and by application to longitudinal data from a fecundity study of fruit flies, for which overdispersion is modeled via a double exponential family.
View details for DOI 10.1111/j.1541-0420.2005.00391.x
View details for Web of Science ID 000236315400021
View details for PubMedID 16542242
-
A new approach to modeling covariate effects and individualization in population pharmacokinetics-pharmacodynamics
JOURNAL OF PHARMACOKINETICS AND PHARMACODYNAMICS
2006; 33 (1): 49-74
Abstract
By combining Laplace's approximation and Monte Carlo methods to evaluate multiple integrals, this paper develops a new approach to estimation in nonlinear mixed effects models that are widely used in population pharmacokinetics and pharmacodynamics. Estimation here involves not only estimating the model parameters from Phase I and II studies but also using the fitted model to estimate the concentration versus time curve or the drug effects of a subject who has covariate information but sparse measurements. Because of its computational tractability, the proposed approach can model the covariate effects nonparametrically by using (i) regression splines or neural networks as basis functions and (ii) AIC or BIC for model selection. Its computational and statistical advantages are illustrated in simulation studies and in Phase I trials.
View details for DOI 10.1007/s10928-005-9000-2
View details for Web of Science ID 000236842900003
View details for PubMedID 16402288
-
Maxima of asymptotically Gaussian random fields and moderate deviation approximations to boundary crossing probabilities of sums of random variables with multidimensional indices
ANNALS OF PROBABILITY
2006; 34 (1): 80-121
View details for Web of Science ID 000235925000003
-
The optimal stopping problem for S-n/n and its ramifications
Festschrift in Honor of Yuan Shih Chow on Random Walk, Sequential Analysis and Related Topics
WORLD SCIENTIFIC PUBL CO PTE LTD. 2006: 131–149
View details for Web of Science ID 000245458600010
-
Autoregressive models with piecewise constant volatility and regression parameters
NBER/NSF Time Series Conference
STATISTICA SINICA. 2005: 279–301
View details for Web of Science ID 000228775700002
-
Optimal stopping for Brownian motion with applications to sequential analysis and option pricing
JOURNAL OF STATISTICAL PLANNING AND INFERENCE
2005; 130 (1-2): 21-47
View details for DOI 10.1016/j.jspi.2003.09.042
View details for Web of Science ID 000226645200003
-
Power, sample size and adaptation considerations in the design of group sequential clinical trials
BIOMETRIKA
2004; 91 (3): 507-528
View details for Web of Science ID 000224077700001
-
Self-normalized processes: Exponential inequalities, moment bounds and iterated logarithm laws
ANNALS OF PROBABILITY
2004; 32 (3A): 1902-1933
View details for DOI 10.1214/009117904000000397
View details for Web of Science ID 000223556800007
-
Exercise regions and efficient valuation of American lookback options
5th ICSA International Statistical Conference
WILEY-BLACKWELL. 2004: 249–69
View details for Web of Science ID 000220357700006
-
Valuation of American options via basis functions
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
2004; 49 (3): 374-385
View details for DOI 10.1109/TAC.2004.824466
View details for Web of Science ID 000220256400006
-
Limit theorems for moving averages
Workshop on Probability wiith Applications to Finance and Insurance
WORLD SCIENTIFIC PUBL CO PTE LTD. 2004: 1–14
View details for Web of Science ID 000229929800001
-
A hybrid estimator in nonlinear and generalised linear mixed effects models
BIOMETRIKA
2003; 90 (4): 859-879
View details for Web of Science ID 000187321500008
-
Saddlepoint approximations and nonlinear boundary crossing probabilities of Markov random walks
ANNALS OF APPLIED PROBABILITY
2003; 13 (2): 395-429
View details for Web of Science ID 000182617800001
-
Nonparametric estimation in nonlinear mixed effects models
BIOMETRIKA
2003; 90 (1): 1-13
View details for Web of Science ID 000181996800001
-
Optimal learning and experimentation in bandit problems
JOURNAL OF ECONOMIC DYNAMICS & CONTROL
2002; 27 (1): 87-108
View details for Web of Science ID 000178224200005
-
Detection and estimation in stochastic systems with time-varying parameters
Workshop on Stochastic Theory and Control
SPRINGER-VERLAG BERLIN. 2002: 251–265
View details for Web of Science ID 000177472100018
-
One-sided tests in clinical trials with multiple endpoints
BIOMETRICS
2001; 57 (4): 1039-1047
Abstract
Treatment comparisons in clinical trials often involve several endpoints. For example, one might wish to demonstrate that a new treatment is superior to the current standard for some components of the multivariate response vector and is not inferior, modulo biologically unimportant difference to the standard treatment for all other components. We introduce a new approach to multiple-endpoint testing that incorporates the essential univariate and multivariate features of the treatment effects. This approach is compared with existing methods in a simulation study and applied to data on rheumatoid arthritis patients receiving one of two treatments.
View details for Web of Science ID 000174956800006
View details for PubMedID 11764242
-
Asymptotic expansions in multidimensional Markov renewal theory and first passage times for Markov random walks
ADVANCES IN APPLIED PROBABILITY
2001; 33 (3): 652-673
View details for Web of Science ID 000172079700008
-
Stochastic neural networks with applications to nonlinear time series
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
2001; 96 (455): 968-981
View details for Web of Science ID 000170729300025
-
Sequential analysis: Some classical problems and new challenges
STATISTICA SINICA
2001; 11 (2): 303-351
View details for Web of Science ID 000168501800001
-
Asymptotic approximations for error probabilities of sequential or fixed sample size tests in exponential families
ANNALS OF STATISTICS
2000; 28 (6): 1638-1669
View details for Web of Science ID 000168627600007
-
Incomplete learning from endogenous data in dynamic allocation
ECONOMETRICA
2000; 68 (6): 1511-1516
View details for Web of Science ID 000090089400008
-
Efficient score estimation and adaptive M-estimators in censored and truncated regression models
STATISTICA SINICA
2000; 10 (3): 731-749
View details for Web of Science ID 000088516900004
-
Sequential multiple hypothesis testing and efficient fault detection-isolation in stochastic systems
IEEE TRANSACTIONS ON INFORMATION THEORY
2000; 46 (2): 595-608
View details for Web of Science ID 000085870000020
-
Hybrid resampling methods for confidence intervals
STATISTICA SINICA
2000; 10 (1): 1-33
View details for Web of Science ID 000085148300001
-
Moment bounds for self-normalized martingales
2nd International Conference on High Dimensional Probability
BIRKHAUSER BOSTON. 2000: 3–11
View details for Web of Science ID 000166793700001
-
Determination of power and sample size in the design of clinical trials with failure-time endpoints and interim analyses
CONTROLLED CLINICAL TRIALS
1999; 20 (5): 423-438
Abstract
An important but difficult task in the design of a clinical trial to compare time to failure between two treatment groups is determination of the number of patients required to achieve a specified power of the test. Because patients typically enter the trial serially and are followed until they fail or withdraw from the study or until the study is terminated, the power of the test depends on the accrual pattern, the noncompliance rate, and the withdrawal rate in addition to the actual survival distributions of the two groups. Incorporating interim analyses and the possibility of early stopping into the trial increases its complexity, and although normal approximations have been developed for computing the significance level of the test when the log-rank or other rank statistics are used, there are no reliable analytic approximations for evaluating the power of the test. This article presents methods, based on Monte Carlo simulations and recent advances in group sequential testing with time-to-event responses, to choose appropriate test statistics, compute power and sample size at specified alternatives, check the adequacy of commonly used normal approximations of the type I error probability, and assess the performance of different interim analysis strategies. It also presents two computer programs implementing these methods.
View details for Web of Science ID 000082659000003
View details for PubMedID 10503802
-
Efficient recursive algorithms for detection of abrupt changes in signals and control systems
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
1999; 44 (5): 952-966
View details for Web of Science ID 000080335000006
-
Regression smoothers and additive models for censored and truncated data
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS
1999; 28 (11): 2717-2747
View details for Web of Science ID 000083512200012
-
Information bounds and quick detection of parameter changes in stochastic systems
IEEE TRANSACTIONS ON INFORMATION THEORY
1998; 44 (7): 2917-2929
View details for Web of Science ID 000078594900011
-
Wald's equations, first passage times and moments of ladder variables in Markov random walks
JOURNAL OF APPLIED PROBABILITY
1998; 35 (3): 566-580
View details for Web of Science ID 000077144200005
-
Resampling methods for confidence intervals in group sequential trials
BIOMETRIKA
1998; 85 (2): 317-332
View details for Web of Science ID 000074556200005
-
Repeated significance testing with censored rank statistics in interim analysis of clinical trials
STATISTICA SINICA
1998; 8 (2): 411-428
View details for Web of Science ID 000073351400008
-
Stochastic adaptive control of linear time-varying systems using auxiliary variables
37th IEEE Conference on Decision and Control
IEEE. 1998: 3445–3450
View details for Web of Science ID 000079708900671
-
Moments of randomly stopped U-STATISTICS
ANNALS OF PROBABILITY
1997; 25 (4): 2055-2081
View details for Web of Science ID 000074004300018
-
Information and prediction criteria for model selection in stochastic regression and ARMA models
STATISTICA SINICA
1997; 7 (2): 285-309
View details for Web of Science ID A1997XD44400008
-
Wald's equation and asymptotic bias of randomly stopped U-statistics
PROCEEDINGS OF THE AMERICAN MATHEMATICAL SOCIETY
1997; 125 (3): 917-925
View details for Web of Science ID A1997WR19600038
-
On optimal stopping problems in sequential hypothesis testing
STATISTICA SINICA
1997; 7 (1): 33-51
View details for Web of Science ID A1997WF57100002
-
Nonparametric estimation and regression analysis with left-truncated and right-censored data
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
1996; 91 (435): 1166-1180
View details for Web of Science ID A1996VK23300026
-
Bootstrap methods for truncated and censored data
STATISTICA SINICA
1996; 6 (3): 509-530
View details for Web of Science ID A1996VB54200001
-
Convergence rate in the strong law of large numbers for Markov chains
Conference on Convergence in Ergodic Theory and Probability
WALTER DE GRUYTER & CO. 1996: 185–192
View details for Web of Science ID A1996BG84W00014
-
Computer-based screening of patients with HIV/AIDS for clinical-trial eligibility.
The Online journal of current clinical trials
1995; Doc No 179: [3347 words, 32 paragraphs]
Abstract
To assess the potential effect of a computer-based system on accrual to clinical trials, we have developed methodology to identify retrospectively and prospectively patients who are eligible or potentially eligible for protocols.Retrospective chart abstraction with computer screening of data for potential protocol eligibility.A county-operated clinic serving human immunodeficiency virus (HIV) positive patients with or without acquired immune deficiency syndrome (AIDS).A randomly selected group of 60 patients who were HIV-infected, 30 of whom had an AIDS-defining diagnosis.Using a computer-based eligibility screening system, for each clinic visit and hospitalization, patients were categorized as eligible, potentially eligible, or ineligible for each of the 17 protocols active during the 7-month study period. Reasons for ineligibility were categorized.None of the patients was enrolled on a clinical trial during the 7-month period. Thirteen patients were identified as eligible for protocol; three patients were eligible for two different protocols; and one patient was eligible for the same protocol during two different time intervals. Fifty-four patients were identified as potentially eligible for a total of 165 accrual opportunities, but important information, such as the result of a required laboratory test, was missing, so that eligibility could not be determined unequivocally. Ineligibility for protocol was determined in 414 (35%) potential opportunities based only on conditions that were amenable to modification, such as the use of concurrent medications; 194 (17%) failed only laboratory tests or subjective determinations not routinely performed; and 346 (29%) failed only routine laboratory tests.There are substantial numbers of eligible and potentially eligible patients who are not enrolled or evaluated for enrollment in prospective clinical trials. Computer-based eligibility screening when coupled with a computer-based medical record offers the potential to identify patients eligible or potentially eligible for clinical trial, to assist in the selection of protocol eligibility criteria, and to make accrual estimates.
View details for PubMedID 7719564
-
ASYMPTOTIC NORMALITY OF A CLASS OF ADAPTIVE STATISTICS WITH APPLICATIONS TO SYNTHETIC DATA METHODS FOR CENSORED REGRESSION
JOURNAL OF MULTIVARIATE ANALYSIS
1995; 52 (2): 259-279
View details for Web of Science ID A1995QH97600004
-
SEQUENTIAL CHANGEPOINT DETECTION IN QUALITY-CONTROL AND DYNAMICAL-SYSTEMS
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-METHODOLOGICAL
1995; 57 (4): 613-658
View details for Web of Science ID A1995RU30700001
-
ASYMPTOTIC PROPERTIES OF NONLINEAR LEAST-SQUARES ESTIMATES IN STOCHASTIC REGRESSION-MODELS
ANNALS OF STATISTICS
1994; 22 (4): 1917-1930
View details for Web of Science ID A1994QX01100019
-
STATISTICAL-ANALYSIS OF LIGAND-BINDING EXPERIMENTS
BIOMETRICS
1994; 50 (3): 782-797
Abstract
After a brief review of commonly used methods for parameter estimation from ligand-binding data in the biochemistry literature, we propose some diagnostic checks and statistical tests of the underlying assumptions and develop methods for evaluating the biases and variances of the estimates and for constructing confidence intervals. Examples on the analysis of data from two radioligand-binding experiments are presented to illustrate these methods.
View details for Web of Science ID A1994PL74800017
View details for PubMedID 7981398
-
A MISSING INFORMATION PRINCIPLE AND M-ESTIMATORS IN REGRESSION-ANALYSIS WITH CENSORED AND TRUNCATED DATA
ANNALS OF STATISTICS
1994; 22 (3): 1222-1255
View details for Web of Science ID A1994QJ59900008
-
Nearly optimal generalized sequential likelihood ratio tests in multivariate exponential families
International Symposium on Multivariate Analysis and Its Applications
INST MATHEMATICAL STATISTICS. 1994: 331–346
View details for Web of Science ID A1994BE60Y00025
-
EDGEWORTH EXPANSIONS FOR SYMMETRICAL STATISTICS WITH APPLICATIONS TO BOOTSTRAP METHODS
STATISTICA SINICA
1993; 3 (2): 517-542
View details for Web of Science ID A1993LY08100016
-
ASYMPTOTIC THEORY OF A BIAS-CORRECTED LEAST-SQUARES ESTIMATOR IN TRUNCATED REGRESSION
STATISTICA SINICA
1992; 2 (2): 519-539
View details for Web of Science ID A1992JQ01200012
-
BOOTSTRAP CONFIDENCE BANDS FOR SPECTRA AND CROSS-SPECTRA
IEEE TRANSACTIONS ON SIGNAL PROCESSING
1992; 40 (5): 1206-1215
View details for Web of Science ID A1992HR45700017
-
ASYMPTOTICALLY EFFICIENT ESTIMATION IN CENSORED AND TRUNCATED REGRESSION-MODELS
STATISTICA SINICA
1992; 2 (1): 17-46
View details for Web of Science ID A1992HC09200002
-
CERTAINTY EQUIVALENCE WITH UNCERTAINTY ADJUSTMENTS IN STOCHASTIC ADAPTIVE-CONTROL
WORKSHOP ON STOCHASTIC THEORY AND ADAPTIVE CONTROL
SPRINGER VERLAG. 1992: 270–284
View details for Web of Science ID A1992LF41800022
-
RECURSIVE-IDENTIFICATION AND ADAPTIVE PREDICTION IN LINEAR STOCHASTIC-SYSTEMS
SIAM JOURNAL ON CONTROL AND OPTIMIZATION
1991; 29 (5): 1061-1090
View details for Web of Science ID A1991GC67900004
-
LARGE SAMPLE THEORY OF A MODIFIED BUCKLEY-JAMES ESTIMATOR FOR REGRESSION-ANALYSIS WITH CENSORED-DATA
ANNALS OF STATISTICS
1991; 19 (3): 1370-1402
View details for Web of Science ID A1991GN04000014
-
PARALLEL RECURSIVE ALGORITHMS IN ASYMPTOTICALLY EFFICIENT ADAPTIVE-CONTROL OF LINEAR STOCHASTIC-SYSTEMS
SIAM JOURNAL ON CONTROL AND OPTIMIZATION
1991; 29 (5): 1091-1127
View details for Web of Science ID A1991GC67900005
-
ADAPTIVE PREDICTION IN NONLINEAR AUTOREGRESSIVE MODELS AND CONTROL-SYSTEMS
STATISTICA SINICA
1991; 1 (2): 309-334
View details for Web of Science ID A1991GA76900001
-
RANK REGRESSION METHODS FOR LEFT-TRUNCATED AND RIGHT-CENSORED DATA
ANNALS OF STATISTICS
1991; 19 (2): 531-556
View details for Web of Science ID A1991FU11400002
-
ESTIMATING A DISTRIBUTION FUNCTION WITH TRUNCATED AND CENSORED-DATA
ANNALS OF STATISTICS
1991; 19 (1): 417-442
View details for Web of Science ID A1991FF04700029
-
INFORMATION BOUNDS, CERTAINTY EQUIVALENCE AND LEARNING IN ASYMPTOTICALLY EFFICIENT ADAPTIVE-CONTROL OF TIME-INVARIANT STOCHASTIC-SYSTEMS
LECTURE NOTES IN CONTROL AND INFORMATION SCIENCES
1991; 161: 335-368
View details for Web of Science ID A1991GF41600015
-
FUNCTIONAL LAWS OF THE ITERATED LOGARITHM FOR THE PRODUCT-LIMIT ESTIMATOR OF A DISTRIBUTION FUNCTION UNDER RANDOM CENSORSHIP OR TRUNCATION
ANNALS OF PROBABILITY
1990; 18 (1): 160-189
View details for Web of Science ID A1990CY73300008
-
STOCHASTIC INTEGRALS OF EMPIRICAL-TYPE PROCESSES WITH APPLICATIONS TO CENSORED REGRESSION
JOURNAL OF MULTIVARIATE ANALYSIS
1988; 27 (2): 334-358
View details for Web of Science ID A1988Q993800002
-
OPEN BANDIT PROCESSES AND OPTIMAL SCHEDULING OF QUEUING-NETWORKS
ADVANCES IN APPLIED PROBABILITY
1988; 20 (2): 447-472
View details for Web of Science ID A1988N875900012
-
NEARLY OPTIMAL SEQUENTIAL-TESTS OF COMPOSITE HYPOTHESES
ANNALS OF STATISTICS
1988; 16 (2): 856-886
View details for Web of Science ID A1988P078900024
-
BOUNDARY CROSSING PROBLEMS FOR SAMPLE MEANS
ANNALS OF PROBABILITY
1988; 16 (1): 375-396
View details for Web of Science ID A1988L925900026
-
HEART-RATE-VARIABILITY DURING RESPIRATORY PAUSES IN PUPPIES AND DOGS
PEDIATRIC RESEARCH
1987; 22 (3): 306-311
Abstract
We studied the time course and change in heart rate during respiratory pauses in puppies (3-4 wk) and young adult dogs. We measured ventilation and ventilatory pattern using barometric plethysmography and recorded the respiratory rate (RR) interval using a pre-processor with an accuracy of 0.2 ms. During tidal breathing, the fluctuations in RR interval were an order of magnitude smaller in the puppy than in the dog. During respiratory pauses in dogs, the RR interval increased sharply, stabilized around the level of expiration of previous breaths, and dropped immediately with the subsequent inspiratory effort. The time course of the change in heart rate was different in the puppy: there was a gradual increase in the RR interval during the entire course of the pause and the maximum RR interval reached was substantially higher than during expiration of previous breaths. Our results suggest that 1) the change in heart rate at the outset of respiratory pauses is too fast to be related to blood gas changes in both puppies and dogs and 2) the mechanisms responsible for the vagal gating of heart rate during tidal breathing and during respiratory pauses are not well developed in early life in the puppy.
View details for Web of Science ID A1987J768700013
View details for PubMedID 3658551
-
DETERMINATION OF SLEEP STATE IN INFANTS USING RESPIRATORY VARIABILITY
PEDIATRIC RESEARCH
1987; 21 (6): 556-562
Abstract
Sleep staging has been conventionally performed using neurophysiologic and behavioral criteria. However, these criteria may not always be available. Since it is known that cardiorespiratory variables in rapid eye movement (REM) sleep are different from those in quiet sleep, we asked whether such variables can be used for the determination of sleep state. We studied nine normal full-term infants at 1 and 4 months of life. Ventilation was measured using barometric plethysmography and the RR interval using a high accuracy R wave detector. Electroencephalogram, electrooculogram, and postural muscle electromyogram were recorded using surface electrodes and behavioral criteria applied. Means of RR interval, respiratory cycle time and tidal volume, and coefficients of variation of the same variables, were obtained for 30-s intervals throughout each sleep study. The Kolmogorov-Smirnov distances between REM and quiet sleep were larger for the coefficients of variation than for the means at both ages for all variables. Moreover, coefficient of variation of respiratory cycle time was found to provide the largest separation between REM and quiet sleep. In view of this result, we developed a statistical decision rule using coefficient of variation of respiratory cycle time for the classification of REM and quiet sleep in blocks of 5-min periods. Each study was divided into 5-min epochs and this rule was applied to each epoch. Of 85 epochs staged as quiet sleep by neurophysiologic and behavioral criteria, 79 epochs (or 93%) were classified correctly as quiet sleep using our decision rule. Of 85 epochs staged as REM sleep, 84 were classified as REM sleep and only one misclassified as quiet sleep.(ABSTRACT TRUNCATED AT 250 WORDS)
View details for Web of Science ID A1987H502600009
View details for PubMedID 3601474
-
ASYMPTOTICALLY EFFICIENT SELF-TUNING REGULATORS
SIAM JOURNAL ON CONTROL AND OPTIMIZATION
1987; 25 (2): 466-481
View details for Web of Science ID A1987G323300012
-
EXTENDED LEAST-SQUARES AND THEIR APPLICATIONS TO ADAPTIVE-CONTROL AND PREDICTION IN LINEAR-SYSTEMS
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
1986; 31 (10): 898-906
View details for Web of Science ID A1986D920600003
-
WITHIN-BREATH ELECTROMYOGRAPHIC CHANGES DURING LOADED BREATHING IN ADULT SHEEP
JOURNAL OF APPLIED PHYSIOLOGY
1986; 61 (4): 1316-1321
Abstract
To investigate the changes in diaphragm electromyogram (EMG) during the course of severe loaded breathing, we subjected five conscious adult sheep to inspiratory flow resistive breathing (resistance greater than 150 cmH2O X l-1 X s) for up to 2-3 h and studied the total EMG power per breath (iEMG) and the EMG power per unit time after dividing the duration of EMG activity within each breath into three equal parts (iEMG1, iEMG2, and iEMG3). Both total breath iEMG and transdiaphragmatic pressure (Pdi) increased, remained at a high level for a certain period of time, and then started to fall. A change in the pattern of iEMG within a breath was observed during loaded breathing. The increase in total-breath iEMG was associated mostly with an increase in iEMG3, or the last part of the EMG power within each inspiration. Similarly, the decrease in total breath iEMG was primarily due to a decrease in iEMG3. We conclude that, in sheep subjected to severe IFR loads for prolonged periods the marked increase in total-breath iEMG at the beginning of loaded breathing and the marked decrease in this iEMG at the time of decrease in Pdi are largely due to changes in iEMG that occur during the latter third of each breath. We speculate that during loaded breathing the recruitment pattern of diaphragmatic muscle fibers changes during the course of an inspiratory effort.
View details for Web of Science ID A1986E439900010
View details for PubMedID 3781946
-
EFFECT OF ENDORPHINS ON HEART-RATE AND BLOOD-PRESSURE IN ADULT DOGS
AMERICAN JOURNAL OF PHYSIOLOGY
1986; 250 (5): H796-H805
Abstract
To investigate the role of opioids in regulating cardiovascular function, we administered delta-opioid receptor agonists D-Ala-D-Leu enkephalin (DADLE) and D-Ala-Met enkephalinamide (DAME), and mu-opioid receptor agonist, a morphiceptin analogue (MA), intracisternally in 13 unanesthetized, chronically instrumented adult dogs in 2 doses (25 and 125 micrograms/kg). After an initial transient drop, the R-R interval increased (peak approximately 25-60 min) postadministration of opioids. The time course and the magnitude of the change in R-R interval depended on the agonist: delta-agonists induced a more prolonged and marked change in R-R interval than mu-agonists at both doses. Mean arterial blood pressure (MAP) increased initially but dropped toward or even below base line 30 min after opioids administration. Atropine, given intravenously or intra-arterially at peak action of agonist in relatively low doses (0.02 mg/kg), induced an AV block followed by a marked decrease in R-R interval. There was also an increase in MAP after atropine. Naloxone, given intracisternally, reversed both delta- and mu-opioid effects but did not induce changes in the R-R interval without prior administration of opioids. We conclude that in unanesthetized adult dogs 1) both mu- and delta-receptor opioid agonists prolong the R-R interval, and this depends on the type of receptor stimulated; 2) opioids induce slowing in heart rate, possibly by increasing parasympathetic activity to the heart; 3) enkephalin and morphiceptin analogues induce a biphasic response in MAP; and 4) endorphins do not modulate cardiovascular function tonically; we speculate that they can alter the R-R interval and MAP in the presence of stimuli.
View details for Web of Science ID A1986C334900015
View details for PubMedID 3518492
-
ASYMPTOTICALLY EFFICIENT ADAPTIVE ALLOCATION RULES
ADVANCES IN APPLIED MATHEMATICS
1985; 6 (1): 4-22
View details for Web of Science ID A1985ADJ5900002
-
ORTHONORMAL BANACH SYSTEMS WITH APPLICATIONS TO LINEAR-PROCESSES
ZEITSCHRIFT FUR WAHRSCHEINLICHKEITSTHEORIE UND VERWANDTE GEBIETE
1985; 70 (3): 381-393
View details for Web of Science ID A1985ART0900007
-
RHYTHMIC VARIATIONS IN R-R INTERVAL DURING SLEEP AND WAKEFULNESS IN PUPPIES AND DOGS
AMERICAN JOURNAL OF PHYSIOLOGY
1984; 247 (1): H67-H73
Abstract
We studied the short-term oscillations in the R-R interval in five puppies at 4 wk of age and five adult dogs during sleep and wakefulness. The R-R interval was measured using an R-R preprocessor, and respiration was recorded using barometric plethysmography. Puppies showed much smaller fluctuations in the R-R interval (SD between 6 and 40 ms) than adult dogs (SD between 124 and 367 ms) in both rapid eye movement (REM) and quiet sleep. Spectral analysis demonstrated that these oscillations were primarily of low frequencies, and the contribution of respiratory sinus arrythmia (RSA) to total power was low. In contrast, in adult dogs during sleep, the spectral distributions were peaked in frequency bands corresponding to mean respiratory rate, and the percent contribution of low frequencies to power was small. Furthermore, the mean R-R interval was considerably larger during expiration than during inspiration in adult dogs (showing 20-140% increase), but not in puppies (showing only -0.4 to 4.4% increase). We conclude that 1) the mechanisms responsible for RSA mature postnatally in the dog; 2) the magnitude of RSA depends on the state of consciousness in the adult dog, being greater in sleep than during wakefulness; and 3) low-frequency oscillations, not related to breathing and independent of sleep state, characterize the variations in the R-R interval in early life but are insignificant in the adult dog.
View details for Web of Science ID A1984TG24800009
View details for PubMedID 6742213
-
OPTIMAL SEQUENTIAL SAMPLING FROM 2 POPULATIONS
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA-PHYSICAL SCIENCES
1984; 81 (4): 1284-1286
Abstract
Given two statistical populations with unknown means, we consider the problem of sampling chi 1, chi 2, ... sequentially from these populations so as to achieve the greatest possible expected value of the sum Sn = chi 1 + ... + chi n. In particular, for normal populations, we obtain the optimal rule and study its properties when the average of the two population means is assumed known, and exhibit an asymptotically optimal rule without assuming any prior knowledge about the population means.
View details for Web of Science ID A1984SH73700006
View details for PubMedID 6583706
-
ADAPTIVE CHOICE OF MEAN OR MEDIAN IN ESTIMATING THE CENTER OF A SYMMETRIC DISTRIBUTION
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA-PHYSICAL SCIENCES
1983; 80 (18): 5803-5806
Abstract
An adaptive choice of the sample mean x(n) or the sample median m(n) is proposed for estimating the center of a symmetric distribution. This choice becomes correct as n --> infinity, and in simulation results for finite n it is almost as good as the better of x(n) and m(n).
View details for Web of Science ID A1983RH33600003
View details for PubMedID 16593376
-
HEART-RATE PATTERN DURING SLEEP IN AN INFANT WITH CONGENITAL PROLONGATION OF THE Q-T INTERVAL (ROMANO-WARD SYNDROME)
CHEST
1983; 84 (2): 191-194
Abstract
Heart rate and the variability of the heart rate, indices of autonomic control, were studied during sleep in an infant with prolonged Q-T interval (Romano-Ward syndrome) and were compared to the heart rate and variability of heart rate in 18 normal infants studied at monthly intervals during the first four months of life. The overall variability and beat-to-beat variability in the infant with Romano-Ward syndrome were significantly below the median in the normal infants at each age and sleep state. This decrease in overall and beat-to-beat variability persisted after normalization by the absolute heart rate; however, the heart rate in the infant with Romano-Ward syndrome was not different from those in normal infants. These data suggest that the presence of a normal heart rate does not exclude abnormal autonomic activity; and in certain clinical situations, the variability of heart rate may be a more sensitive index of abnormal autonomic function than the heart rate itself.
View details for Web of Science ID A1983RB41500018
View details for PubMedID 6872600
-
ENKEPHALIN-INDUCED CHANGES IN VENTILATION AND VENTILATORY PATTERN IN ADULT DOGS
JOURNAL OF APPLIED PHYSIOLOGY
1983; 55 (4): 1311-1320
Abstract
We studied the changes in ventilation induced by intracisternal administration of enkephalins in four unanesthetized adult dogs. Instantaneous minute ventilation (VT/TT) decreased markedly after D-Ala-Met-enkephalinamide (DAME). Mean VT/TT decreased maximally by 20-50 min after DAME and lasted an additional 15-60 min; by 2 h, VT/TT had returned to base line. Four doses (5, 25, 60, and 125 micrograms/kg) of DAME were used, and the ventilatory response depended on the dose. Mean inspiratory time decreased but mean expiratory time and mean TT showed a marked prolongation. Periodic breathing (2-3 breaths separated by long apneic pauses) occurred in every study and the frequency of sighs increased considerably. All these ventilatory changes were reversed by low doses of naloxone or naltrexone; in addition, VT/TT increased well above base line after the administration of these antagonists. However, naloxone did not increase VT/TT when injected without prior administration of DAME. We conclude that 1) the decrease in VT/TT is due to a decrease in respiratory duty cycle; 2) periodic breathing and increased frequency of sighs constitute part of the changes in the ventilatory pattern induced by DAME; 3) a ventilatory withdrawal reaction may occur after a receptor-agonist interaction of short duration; and 4) although enkephalins can modulate ventilation and the breathing pattern in a major way, these data provide no evidence suggesting that this modulation is tonic.
View details for Web of Science ID A1983RL96500040
View details for PubMedID 6629966
-
FIXED ACCURACY ESTIMATION OF AN AUTOREGRESSIVE PARAMETER
ANNALS OF STATISTICS
1983; 11 (2): 478-485
View details for Web of Science ID A1983QT81500010
-
CONVERGENCE SYSTEMS AND STRONG CONSISTENCY OF LEAST-SQUARES ESTIMATES IN REGRESSION-MODELS
KEXUE TONGBAO
1983; 28 (1): 16-20
View details for Web of Science ID A1983QF00600004
-
ASYMPTOTIC PROPERTIES OF GENERAL AUTOREGRESSIVE MODELS AND STRONG CONSISTENCY OF LEAST-SQUARES ESTIMATES OF THEIR PARAMETERS
JOURNAL OF MULTIVARIATE ANALYSIS
1983; 13 (1): 1-23
View details for Web of Science ID A1983QN13200001
-
LEAST-SQUARES ESTIMATES IN STOCHASTIC REGRESSION-MODELS WITH APPLICATIONS TO IDENTIFICATION AND CONTROL OF DYNAMIC-SYSTEMS
ANNALS OF STATISTICS
1982; 10 (1): 154-166
View details for Web of Science ID A1982NG05200011
-
DETERMINATION OF VENTILATORY PATTERN IN REM-SLEEP IN NORMAL INFANTS
JOURNAL OF APPLIED PHYSIOLOGY
1982; 53 (1): 52-56
Abstract
Using methods that we devised for detecting and counting eye movements on the electrooculogram (EOG), we studied tidal volume (VT) and total respiratory cycle time (Ttot) as a function of the frequency of rapid eye movements (REM) during REM sleep in nine normal infants at 1 mo of age. In each of the nine infants, the mean VT and mean Ttot decreased with increasing frequency of eye movements. Instantaneous minute ventilation (VT/Ttot or V), however, did not change with the frequency of eye movements. In addition, there was no consistent change in the variability of VT, Ttot, or V when studied as a function of the frequency of eye movements. Our data support the notion that the ventilatory pattern in REM sleep depends in part on mechanisms that are inherent to REM sleep.
View details for Web of Science ID A1982NX42800008
View details for PubMedID 7118647
-
BREATH-TO-BREATH VARIATIONS IN RATE AND DEPTH OF VENTILATION IN SLEEPING INFANTS
AMERICAN JOURNAL OF PHYSIOLOGY
1982; 243 (1): R164-R169
Abstract
Ventilatory measurements were made noninvasively over 2- to 3-h periods during sleep in each of nine normal infants at 1 mo of age. To assess the changes that occur in ventilation on a breath-to-breath basis, we 1) examined the variations of each of tidal volume (VT), respiratory cycle time (Ttot), expiratory time (TE), and inspiratory time (TI) and 2) studied their interrelationships. We found that the variations of VT, Ttot, and TE but not of TI were significantly greater in rapid-eye-movement (REM) than in quiet sleep. In addition, on a breath-to-breath basis, VT had a positive linear relationship and strong correlation with TI; however, the correlation between VT and TE was weak in both sleep states. VT/Ttot was found to be moderately and negatively correlated with Ttot in both REM and quiet sleep. VT was weakly correlated with Ttot in REM sleep and was, on the average, more correlated with Ttot in quiet sleep. We suggest that in infants 1) on a breath-to-breath basis, VT/Ttot is likely to drop if respiratory frequency is decreased and 2) VT is nonlinearly related to Ttot during sleep; this lack of linearity depends on the lack of constancy of VT/Ttot, which is in turn closely related to the variability of the "on-switching" of inspiratory activity.
View details for Web of Science ID A1982NX81500079
View details for PubMedID 7091389
-
ASYMPTOTIC PROPERTIES OF PROJECTIONS WITH APPLICATIONS TO STOCHASTIC REGRESSION PROBLEMS
JOURNAL OF MULTIVARIATE ANALYSIS
1982; 12 (3): 346-370
View details for Web of Science ID A1982PC01000003
-
A LAW OF THE ITERATED LOGARITHM FOR DOUBLE ARRAYS OF INDEPENDENT RANDOM-VARIABLES WITH APPLICATIONS TO REGRESSION AND TIME-SERIES MODELS
ANNALS OF PROBABILITY
1982; 10 (2): 320-335
View details for Web of Science ID A1982NN94800004
-
CONSISTENCY AND ASYMPTOTIC EFFICIENCY OF SLOPE ESTIMATES IN STOCHASTIC-APPROXIMATION SCHEMES
ZEITSCHRIFT FUR WAHRSCHEINLICHKEITSTHEORIE UND VERWANDTE GEBIETE
1981; 56 (3): 329-360
View details for Web of Science ID A1981LX07900003
-
CONVERGENCE SYSTEMS AND STRONG CONSISTENCY OF LEAST-SQUARES ESTIMATES IN REGRESSION-MODELS
JOURNAL OF MULTIVARIATE ANALYSIS
1981; 11 (3): 319-333
View details for Web of Science ID A1981MF62300002
-
VENTILATION AND VENTILATORY PATTERN DURING SLEEP IN ABORTED SUDDEN INFANT DEATH SYNDROME
PEDIATRIC RESEARCH
1981; 15 (5): 879-883
Abstract
To assess ventilatory control during sleep in infants at risk for the sudden infant death syndrome (SIDS), we made serial measurements of resting tidal volume (Vt), respiratory cycle time (Ttot), and the ventilatory changes resulting from inhalation of 2% CO2 in aborted SIDS infants in rapid eye movement and quiet sleep and compared them to a group of normal infants during the first 4 months of life. Ventilation was measured by the barometric method, and sleep was staged using electroencephalogram, electrooculogram, and electromyogram and behavioral criteria. Although resting instantaneous minute ventilation (Vt/Ttot) was virtually the same in both groups of infants, Vt tended to be smaller (by up to 50% in the first 2 months) and Ttot tended to be shorter in aborted SIDS than in normal infants in both rapid eye movement and quiet sleep. The increase in the mean Vt/Ttot with 2% CO2 is greater by about 5 to 20% in aborted SIDS than in normal infants at 3 and 4 months of age in both sleep states. These findings, together with our previous findings that aborted SIDS infants have an increase in heart rate and a shortening of the QT interval, provide indirect evidence that infants at high risk for SIDS may have increased sympathoadrenal activity.
View details for Web of Science ID A1981LM90000020
View details for PubMedID 6787545
-
SEQUENTIAL MEDICAL TRIALS
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA-PHYSICAL SCIENCES
1980; 77 (6): 3135-3138
Abstract
A model for sequential clinical trials is discussed. Three proposed stopping rules are studied by the Monte Carlo method for small patient horizons and mathematically for large patient horizons. They are shown to be about equally effective and asymptotically optimal from both Bayesian and frequentist points of view and are markedly superior to any fixed sample size procedure.
View details for Web of Science ID A1980JY17300016
View details for PubMedID 16592839
-
HEART-RATE AND HEART-RATE-VARIABILITY DURING SLEEP IN ABORTED SUDDEN INFANT DEATH SYNDROME
JOURNAL OF PEDIATRICS
1980; 97 (1): 51-55
Abstract
Heart rate and heart rate variability were studied during sleep at monthly intervals in 18 normal infants and 12 infants with aborted sudden infant death syndrome during the first four months of life. At each age studied and in both REM and quiet sleep, the aborted SIDS infants had a 5 to 10% faster heart rate. Moreover, the aborted SIDS infants had a 10 to 45% smaller beat-to-beat and overall heart rate variability. Although the differences in overall variability persisted after normalization by the absolute heart rate, the differences in the beat-to-beat variability narrowed. These findings, when taken in conjunction with our previous observation that aborted SIDS infants have a smaller QT index than normal infants, suggest that infants with aborted SIDS have an increase in sympathetic activity or in circulating levels of catecholamines.
View details for Web of Science ID A1980JY86700010
View details for PubMedID 7381648
-
NON-LINEAR RENEWAL THEORY WITH APPLICATIONS TO SEQUENTIAL-ANALYSIS .2.
ANNALS OF STATISTICS
1979; 7 (1): 60-76
View details for Web of Science ID A1979GL00300004
-
ADAPTIVE DESIGN AND STOCHASTIC-APPROXIMATION
ANNALS OF STATISTICS
1979; 7 (6): 1196-1221
View details for Web of Science ID A1979HW07100004
-
STRONG CONSISTENCY OF LEAST-SQUARES ESTIMATES IN MULTIPLE-REGRESSION .2.
JOURNAL OF MULTIVARIATE ANALYSIS
1979; 9 (3): 343-361
View details for Web of Science ID A1979HW07400001
-
LOCAL CONVERGENCE THEOREMS FOR ADAPTIVE STOCHASTIC-APPROXIMATION SCHEMES
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
1979; 76 (7): 3065-3067
Abstract
For the regression model y = M(x) + epsilon, adaptive stochastic approximation schemes of the form x(n+1) = x(n) - y(n)/(nb(n)) for choosing the levels x(1),x(2),... at which y(1),y(2),... are observed converge with probability 1 to the unknown root theta of the regression function M(x). Certain local convergence theorems that relate the convergence rate of x(n) - theta to the limiting behavior of the random variables b(n) are established.
View details for Web of Science ID A1979HE02600007
View details for PubMedID 16592673
-
CLASS OF DEPENDENT RANDOM-VARIABLES AND THEIR MAXIMA
ZEITSCHRIFT FUR WAHRSCHEINLICHKEITSTHEORIE UND VERWANDTE GEBIETE
1978; 42 (2): 89-111
View details for Web of Science ID A1978EQ52900001
-
STRONG CONSISTENCY OF LEAST-SQUARES ESTIMATES IN MULTIPLE-REGRESSION
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
1978; 75 (7): 3034-3036
Abstract
The strong consistency of least squares estimates in multiple regression models with independent errors is obtained under minimal assumptions on the design and weak moment conditions on the errors.
View details for Web of Science ID A1978FJ88100007
View details for PubMedID 16592540
-
LIMIT-THEOREMS FOR WEIGHTED SUMS AND STOCHASTIC-APPROXIMATION PROCESSES
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
1978; 75 (3): 1068-1070
Abstract
By establishing certain limit theorems for martingale transforms and weighted sums, a Marcinkiewicz-Zygmund strong law together with weak and strong invariance principles is developed for stochastic approximation processes, under minimal conditions on the errors.
View details for Web of Science ID A1978EU94600008
View details for PubMedID 16592509
-
ADAPTIVE DESIGN IN REGRESSION AND CONTROL
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
1978; 75 (2): 586-587
Abstract
When y = M(x) + epsilon, where M may be nonlinear, adaptive regression designs of the levels x(1), x(2),... at which y(1), y(2),... are observed lead to asymptotically efficient estimates of the value theta of x for which M(theta) is equal to any desired value y(*). More importantly, these designs also make the "cost" of the observations, defined at the nth stage to be Sigma(1) (n) (x(i) - theta)(2), to be of the order of log n instead of n, an obvious advantage in medical and other applications.
View details for Web of Science ID A1978EQ77900013
View details for PubMedID 16592494
-
NONLINEAR RENEWAL THEORY WITH APPLICATIONS TO SEQUENTIAL-ANALYSIS I
ANNALS OF STATISTICS
1977; 5 (5): 946-954
View details for Web of Science ID A1977DV19800010
-
STRONG CONSISTENCY OF LEAST-SQUARES ESTIMATES IN REGRESSION-MODELS
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
1977; 74 (7): 2667-2669
Abstract
A general theorem on the limiting behavior of certain weighted sums of i.i.d. random variables is obtained. This theorem is then applied to prove the strong consistency of least-squares estimates in linear and nonlinear regression models with i.i.d. errors under minimal assumptions on the design and weak moment conditions on the errors.
View details for Web of Science ID A1977DP86600020
View details for PubMedID 16592416
-
MAXIMALLY DEPENDENT RANDOM-VARIABLES
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
1976; 73 (2): 286-288
Abstract
Let X(1),..., X(n) have an arbitrary common marginal distribution function F, and let M(n) = max(X(1),..., X(n)). It is shown that EM(n) = m(n), where m(n) = a(n) + n[unk](an) (infinity)[1 -F(x)]dx and = F(-1)(1 - n(-1)), and that EM(n) = m(n) when X(1),..., X(n) are "maximally dependent"; i.e., P(M(n) > x) = min{1, n[1 - F(x)]} for all x. Moreover, as n --> infinity, a(n) approximately m(n) approximately m(n) (*), where m(n) (*) = EM(n) when X(1),..., X(n) are independent, provided that [1 - F(cx)]/[1 - F(x)] --> 0 as x --> infinity for every c > 1, and E(X(1) (-))(r) < infinity for some r > 0. The case in which F is standard normal is considered in detail.
View details for Web of Science ID A1976BG66500007
View details for PubMedID 16578739