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.

Academic Appointments


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


All Publications


  • Power of an Adaptive Trial Design for Endovascular Stroke Studies Simulations Using IMS (Interventional Management of Stroke) III Data STROKE Lansberg, M. G., Bhat, N. S., Yeatts, S. D., Palesch, Y. Y., Broderick, J. P., Albers, G. W., Lai, T. L., Lavori, P. W. 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 Lai, T. L., Tsang, K. W. 2016; 26 (4): 1411-1426
  • Discussion on "Sequential detection/isolation of abrupt changes" by Igor V. Nikiforov SEQUENTIAL ANALYSIS-DESIGN METHODS AND APPLICATIONS Lai, T. L., Tsang, K. W. 2016; 35 (3): 305-310
  • Innovative designs of point-of-care comparative effectiveness trials CONTEMPORARY CLINICAL TRIALS Shih, M., Turakhia, M., Lai, T. L. 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

  • Adaptive design of confirmatory trials: Advances and challenges CONTEMPORARY CLINICAL TRIALS Lai, T. L., Lavori, P. W., Tsang, K. W. 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

  • ABO Mismatch Is Associated with Increased Nonrelapse Mortality after Allogeneic Hematopoietic Cell Transplantation BIOLOGY OF BLOOD AND MARROW TRANSPLANTATION Logan, A. C., Wang, Z., Alimoghaddam, K., Wong, R. M., Lai, T., Negrin, R. S., Grumet, C., Logan, B. R., Zhang, M., Spellman, S. R., Lee, S. J., Miklos, D. B. 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 Lai, T. L., Lim, J. 2015; 25 (1): 403-421
  • DYNAMIC EMPIRICAL BAYES MODELS AND THEIR APPLICATIONS TO LONGITUDINAL DATA ANALYSIS AND PREDICTION STATISTICA SINICA Lai, T. L., Su, Y., Sun, K. H. 2014; 24 (4): 1505-1528
  • Discussion on "Sequential Estimation for Time Series Models" by T. N. Sriram and Ross Iaci SEQUENTIAL ANALYSIS-DESIGN METHODS AND APPLICATIONS Lai, T. L. 2014; 33 (2): 169-173
  • A GENERAL THEORY OF PARTICLE FILTERS IN HIDDEN MARKOV MODELS AND SOME APPLICATIONS ANNALS OF STATISTICS Chan, H. P., Lai, T. L. 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 Lai, T. L., Liao, O. Y., Kim, D. W. 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 Lai, T. L., Liao, O. Y., Kim, D. W. 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 Lai, T. L., Xing, H. 2013; 23 (4): 1573-1594
  • RARE-EVENT SIMULATION OF HEAVY-TAILED RANDOM WALKS BY SEQUENTIAL IMPORTANCE SAMPLING AND RESAMPLING ADVANCES IN APPLIED PROBABILITY Chan, H. P., Deng, S., Lai, T. 2012; 44 (4): 1173-1196
  • Sequential design of phase II-III cancer trials STATISTICS IN MEDICINE Lai, T. L., Lavori, P. W., Shih, M. 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 Lai, T. L., Lavori, P. W., Shih, M. I., Sikic, B. I. 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 Lai, T. L., Lavori, P. W., Shih, M. 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 Lai, T. L., Liao, O. Y., Zhu, R. G. 2012; 5 (4): 431-442
  • Sequential Importance Sampling and Resampling for Dynamic Portfolio Credit Risk OPERATIONS RESEARCH Deng, S., Giesecke, K., Lai, T. L. 2012; 60 (1): 78-91
  • Futility stopping in clinical trials STATISTICS AND ITS INTERFACE He, P., Lai, T. L., Liao, O. Y. 2012; 5 (4): 415-423
  • Efficient Adaptive Randomization and Stopping Rules in Multi-arm Clinical Trials for Testing a New Treatment 3rd International Workshop on Sequential Methodologies (IWSM) Lai, T. L., Liao, O. Y. TAYLOR & FRANCIS INC. 2012: 441–57
  • A SEQUENTIAL MONTE CARLO APPROACH TO COMPUTING TAIL PROBABILITIES IN STOCHASTIC MODELS ANNALS OF APPLIED PROBABILITY Chan, H. P., Lai, T. L. 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 Lai, T. L., Gross, S. T., Shen, D. B. 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 Ing, C., Lai, T. L. 2011; 21 (4): 1473-1513
  • MEAN-VARIANCE PORTFOLIO OPTIMIZATION WHEN MEANS AND COVARIANCES ARE UNKNOWN ANNALS OF APPLIED STATISTICS Lai, T. L., Xing, H., Chen, Z. 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 Bartroff, J., Lai, T. L. 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 Lai, T. L., Xing, H. 2011; 21 (2): 539-569
  • CRAMER TYPE MODERATE DEVIATIONS FOR STUDENTIZED U-STATISTICS ESAIM-PROBABILITY AND STATISTICS Lai, T. L., Shao, Q., Wang, Q. 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 Ing, C., Lai, T. L. 2011; 30 (4): 404-411
  • Sequential generalized likelihood ratio tests for vaccine safety evaluation STATISTICS IN MEDICINE Shih, M., Lai, T. L., Heyse, J. F., Chen, J. 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 Bartroff, J., Lai, T. L. 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 Bartroff, J., Lai, T. L. 2010; 39 (8-9): 1597-1607
  • Theory and applications of multivariate self-normalized processes STOCHASTIC PROCESSES AND THEIR APPLICATIONS de la Pena, V. H., Klass, M. J., Lai, T. L. 2009; 119 (12): 4210-4227
  • Option hedging theory under transaction costs JOURNAL OF ECONOMIC DYNAMICS & CONTROL Lai, T. L., Lim, T. W. 2009; 33 (12): 1945-1961
  • Tests and confidence intervals for secondary endpoints in sequential clinical trials BIOMETRIKA Lai, T. L., Shih, M., Su, Z. 2009; 96 (4): 903-915
  • A Bayesian Approach to Sequential Surveillance in Exponential Families COMMUNICATIONS IN STATISTICS-THEORY AND METHODS Lai, T. L., Liu, T., Xing, H. 2009; 38 (16-17): 2958-2968
  • Modern sequential analysis and its applications to computerized adaptive testing PSYCHOMETRIKA Bartroff, J., Finkelman, M., Lai, T. L. 2008; 73 (3): 473-486
  • Statistical inference in dynamic panel data models JOURNAL OF STATISTICAL PLANNING AND INFERENCE Lai, T. L., Small, D. S., Liu, J. 2008; 138 (9): 2763-2776
  • Stochastic segmentation models for array-based comparative genomic hybridization data analysis BIOSTATISTICS Lai, T. L., Xing, H., Zhang, N. 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 Lai, T. L., Wong, S. P. 2008; 1 (2): 229-241
  • A Hidden Markov Filtering Approach to Multiple Change-point Models 47th IEEE Conference on Decision and Control Lai, T. L., Xing, H. IEEE. 2008: 1914–1919
  • Corrected random walk approximations to free boundary problems in optimal stopping ADVANCES IN APPLIED PROBABILITY Lai, T. L., Yao, Y., Aitsahlia, F. 2007; 39 (3): 753-775
  • Efficient importance sampling for Monte Carlo evaluation of exceedance probabilities ANNALS OF APPLIED PROBABILITY Chan, H. P., Lai, T. L. 2007; 17 (2): 440-473
  • A combined superiority and non-inferiority approach to multiple endpoints in clinical trials STATISTICS IN MEDICINE Bloch, D. A., Lai, T. L., Su, Z., Tubert-Bitter, P. 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 Lai, T. L., Xing, H. 2007; 137 (3): 1066-1075
  • 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 Lai, T. L., Small, D. 2007; 69: 79-99
  • Identification and adaptive control of change-point ARX models via Rao-Blackwellized particle filters IEEE TRANSACTIONS ON AUTOMATIC CONTROL Chen, Y., Lai, T. L. 2007; 52 (1): 67-72
  • Confidence intervals for survival quantiles in the Cox regression model LIFETIME DATA ANALYSIS Lai, T. L., Su, Z. 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 Lai, T. L., Li, W. 2006; 93 (3): 641-654
  • Efficient recursive estimation and adaptive control in stochastic regression and ARMAX models STATISTICA SINICA Lai, T. L., Ying, Z. 2006; 16 (3): 741-772
  • Modified Haybittle-Peto group sequential designs for testing superiority and non-inferiority hypotheses in clinical trials STATISTICS IN MEDICINE Lai, T. L., Shih, M. C., Zhu, G. R. 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 Lai, T. L., Shih, M. C., Wong, S. P. 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 Lai, T. L., Shih, M. C., Wong, S. P. 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 Chan, H. P., Lai, T. L. 2006; 34 (1): 80-121
  • 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 Lai, T. L., Yao, Y. WORLD SCIENTIFIC PUBL CO PTE LTD. 2006: 131–149
  • Autoregressive models with piecewise constant volatility and regression parameters NBER/NSF Time Series Conference Lai, T. L., Liu, H. Y., Xing, H. P. STATISTICA SINICA. 2005: 279–301
  • Optimal stopping for Brownian motion with applications to sequential analysis and option pricing JOURNAL OF STATISTICAL PLANNING AND INFERENCE Lai, T. L., Lim, T. W. 2005; 130 (1-2): 21-47
  • Power, sample size and adaptation considerations in the design of group sequential clinical trials BIOMETRIKA Lai, T. L., Shih, M. C. 2004; 91 (3): 507-528
  • Self-normalized processes: Exponential inequalities, moment bounds and iterated logarithm laws ANNALS OF PROBABILITY De La Pena, V. H., Klass, M. J., Lai, T. L. 2004; 32 (3A): 1902-1933
  • Exercise regions and efficient valuation of American lookback options 5th ICSA International Statistical Conference Lai, T. L., Lim, T. W. WILEY-BLACKWELL. 2004: 249–69
  • Valuation of American options via basis functions IEEE TRANSACTIONS ON AUTOMATIC CONTROL Lai, T. L., Wong, S. P. 2004; 49 (3): 374-385
  • Limit theorems for moving averages Workshop on Probability wiith Applications to Finance and Insurance Lai, T. L. WORLD SCIENTIFIC PUBL CO PTE LTD. 2004: 1–14
  • A hybrid estimator in nonlinear and generalised linear mixed effects models BIOMETRIKA Lai, T. L., Shih, M. C. 2003; 90 (4): 859-879
  • Saddlepoint approximations and nonlinear boundary crossing probabilities of Markov random walks ANNALS OF APPLIED PROBABILITY Chan, H. P., Lai, T. L. 2003; 13 (2): 395-429
  • Nonparametric estimation in nonlinear mixed effects models BIOMETRIKA Lai, T. L., Shih, M. C. 2003; 90 (1): 1-13
  • Optimal learning and experimentation in bandit problems JOURNAL OF ECONOMIC DYNAMICS & CONTROL Brezzi, M., Lai, T. L. 2002; 27 (1): 87-108
  • Detection and estimation in stochastic systems with time-varying parameters Workshop on Stochastic Theory and Control Lai, T. L. SPRINGER-VERLAG BERLIN. 2002: 251–265
  • One-sided tests in clinical trials with multiple endpoints BIOMETRICS Bloch, D. A., Lai, T. L., Tubert-Bitter, P. 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

  • Stochastic neural networks with applications to nonlinear time series JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION Lai, T. L., Wong, S. P. 2001; 96 (455): 968-981
  • Asymptotic expansions in multidimensional Markov renewal theory and first passage times for Markov random walks ADVANCES IN APPLIED PROBABILITY Fuh, C. D., Lai, T. L. 2001; 33 (3): 652-673
  • Sequential analysis: Some classical problems and new challenges STATISTICA SINICA Lai, T. L. 2001; 11 (2): 303-351
  • Asymptotic approximations for error probabilities of sequential or fixed sample size tests in exponential families ANNALS OF STATISTICS Chan, H. P., Lai, T. L. 2000; 28 (6): 1638-1669
  • Incomplete learning from endogenous data in dynamic allocation ECONOMETRICA Brezzi, M., Lai, T. L. 2000; 68 (6): 1511-1516
  • Efficient score estimation and adaptive M-estimators in censored and truncated regression models STATISTICA SINICA Kim, C. K., Lai, T. L. 2000; 10 (3): 731-749
  • Sequential multiple hypothesis testing and efficient fault detection-isolation in stochastic systems IEEE TRANSACTIONS ON INFORMATION THEORY Lai, T. L. 2000; 46 (2): 595-608
  • Hybrid resampling methods for confidence intervals STATISTICA SINICA Chuang, C. S., Lai, T. L. 2000; 10 (1): 1-33
  • Moment bounds for self-normalized martingales 2nd International Conference on High Dimensional Probability de la Pena, V. H., Klass, M. J., Lai, T. L. BIRKHAUSER BOSTON. 2000: 3–11
  • Determination of power and sample size in the design of clinical trials with failure-time endpoints and interim analyses CONTROLLED CLINICAL TRIALS Gu, M. G., Lai, T. L. 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 Lai, T. L., Shan, J. Z. 1999; 44 (5): 952-966
  • Regression smoothers and additive models for censored and truncated data COMMUNICATIONS IN STATISTICS-THEORY AND METHODS Kim, C. K., Lai, T. L. 1999; 28 (11): 2717-2747
  • Information bounds and quick detection of parameter changes in stochastic systems IEEE TRANSACTIONS ON INFORMATION THEORY Lai, T. L. 1998; 44 (7): 2917-2929
  • Wald's equations, first passage times and moments of ladder variables in Markov random walks JOURNAL OF APPLIED PROBABILITY Fuh, C. D., Lai, T. L. 1998; 35 (3): 566-580
  • Resampling methods for confidence intervals in group sequential trials BIOMETRIKA Chuang, C. S., Lai, T. L. 1998; 85 (2): 317-332
  • Repeated significance testing with censored rank statistics in interim analysis of clinical trials STATISTICA SINICA Gu, M. G., Lai, T. L. 1998; 8 (2): 411-428
  • Stochastic adaptive control of linear time-varying systems using auxiliary variables 37th IEEE Conference on Decision and Control Lai, T. L., Li, Z. IEEE. 1998: 3445–3450
  • Moments of randomly stopped U-STATISTICS ANNALS OF PROBABILITY de la Pena, V. H., Lai, T. L. 1997; 25 (4): 2055-2081
  • Information and prediction criteria for model selection in stochastic regression and ARMA models STATISTICA SINICA Lai, T. L., Lee, C. P. 1997; 7 (2): 285-309
  • Wald's equation and asymptotic bias of randomly stopped U-statistics PROCEEDINGS OF THE AMERICAN MATHEMATICAL SOCIETY DELAPENA, V. H., Lai, T. L. 1997; 125 (3): 917-925
  • On optimal stopping problems in sequential hypothesis testing STATISTICA SINICA Lai, T. L. 1997; 7 (1): 33-51
  • Nonparametric estimation and regression analysis with left-truncated and right-censored data JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION Gross, S. T., Lai, T. L. 1996; 91 (435): 1166-1180
  • Bootstrap methods for truncated and censored data STATISTICA SINICA Gross, S. T., Lai, T. L. 1996; 6 (3): 509-530
  • Convergence rate in the strong law of large numbers for Markov chains Conference on Convergence in Ergodic Theory and Probability Fuh, C. D., Lai, T. L. WALTER DE GRUYTER & CO. 1996: 185–192
  • Computer-based screening of patients with HIV/AIDS for clinical-trial eligibility. The Online journal of current clinical trials Carlson, R. W., Tu, S. W., Lane, N. M., Lai, T. L., Kemper, C. A., Musen, M. A., Shortliffe, E. H. 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 Lai, T. L., Ying, Z. L., Zheng, Z. K. 1995; 52 (2): 259-279
  • SEQUENTIAL CHANGEPOINT DETECTION IN QUALITY-CONTROL AND DYNAMICAL-SYSTEMS JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-METHODOLOGICAL Lai, T. L. 1995; 57 (4): 613-658
  • ASYMPTOTIC PROPERTIES OF NONLINEAR LEAST-SQUARES ESTIMATES IN STOCHASTIC REGRESSION-MODELS ANNALS OF STATISTICS Lai, T. L. 1994; 22 (4): 1917-1930
  • STATISTICAL-ANALYSIS OF LIGAND-BINDING EXPERIMENTS BIOMETRICS Lai, T. L., Zhang, L. M. 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 Lai, T. L., Ying, Z. L. 1994; 22 (3): 1222-1255
  • Nearly optimal generalized sequential likelihood ratio tests in multivariate exponential families International Symposium on Multivariate Analysis and Its Applications Lai, T. L., Zhang, L. M. INST MATHEMATICAL STATISTICS. 1994: 331–346
  • EDGEWORTH EXPANSIONS FOR SYMMETRICAL STATISTICS WITH APPLICATIONS TO BOOTSTRAP METHODS STATISTICA SINICA Lai, T. L., WANG, J. Q. 1993; 3 (2): 517-542
  • ASYMPTOTIC THEORY OF A BIAS-CORRECTED LEAST-SQUARES ESTIMATOR IN TRUNCATED REGRESSION STATISTICA SINICA Lai, T. L., Ying, Z. 1992; 2 (2): 519-539
  • BOOTSTRAP CONFIDENCE BANDS FOR SPECTRA AND CROSS-SPECTRA IEEE TRANSACTIONS ON SIGNAL PROCESSING Politis, D. N., Romano, J. P., Lai, T. L. 1992; 40 (5): 1206-1215
  • ASYMPTOTICALLY EFFICIENT ESTIMATION IN CENSORED AND TRUNCATED REGRESSION-MODELS STATISTICA SINICA Lai, T. L., Ying, Z. L. 1992; 2 (1): 17-46
  • CERTAINTY EQUIVALENCE WITH UNCERTAINTY ADJUSTMENTS IN STOCHASTIC ADAPTIVE-CONTROL WORKSHOP ON STOCHASTIC THEORY AND ADAPTIVE CONTROL Lai, T. L. SPRINGER VERLAG. 1992: 270–284
  • RECURSIVE-IDENTIFICATION AND ADAPTIVE PREDICTION IN LINEAR STOCHASTIC-SYSTEMS SIAM JOURNAL ON CONTROL AND OPTIMIZATION Lai, T. L., Ying, Z. L. 1991; 29 (5): 1061-1090
  • LARGE SAMPLE THEORY OF A MODIFIED BUCKLEY-JAMES ESTIMATOR FOR REGRESSION-ANALYSIS WITH CENSORED-DATA ANNALS OF STATISTICS Lai, T. L., Ying, Z. L. 1991; 19 (3): 1370-1402
  • PARALLEL RECURSIVE ALGORITHMS IN ASYMPTOTICALLY EFFICIENT ADAPTIVE-CONTROL OF LINEAR STOCHASTIC-SYSTEMS SIAM JOURNAL ON CONTROL AND OPTIMIZATION Lai, T. L., Ying, Z. L. 1991; 29 (5): 1091-1127
  • ADAPTIVE PREDICTION IN NONLINEAR AUTOREGRESSIVE MODELS AND CONTROL-SYSTEMS STATISTICA SINICA Lai, T. L., Zhu, G. G. 1991; 1 (2): 309-334
  • RANK REGRESSION METHODS FOR LEFT-TRUNCATED AND RIGHT-CENSORED DATA ANNALS OF STATISTICS Lai, T. L., Ying, Z. L. 1991; 19 (2): 531-556
  • ESTIMATING A DISTRIBUTION FUNCTION WITH TRUNCATED AND CENSORED-DATA ANNALS OF STATISTICS Lai, T. L., Ying, Z. L. 1991; 19 (1): 417-442
  • INFORMATION BOUNDS, CERTAINTY EQUIVALENCE AND LEARNING IN ASYMPTOTICALLY EFFICIENT ADAPTIVE-CONTROL OF TIME-INVARIANT STOCHASTIC-SYSTEMS LECTURE NOTES IN CONTROL AND INFORMATION SCIENCES Lai, T. L. 1991; 161: 335-368
  • FUNCTIONAL LAWS OF THE ITERATED LOGARITHM FOR THE PRODUCT-LIMIT ESTIMATOR OF A DISTRIBUTION FUNCTION UNDER RANDOM CENSORSHIP OR TRUNCATION ANNALS OF PROBABILITY Gu, M. G., Lai, T. L. 1990; 18 (1): 160-189
  • STOCHASTIC INTEGRALS OF EMPIRICAL-TYPE PROCESSES WITH APPLICATIONS TO CENSORED REGRESSION JOURNAL OF MULTIVARIATE ANALYSIS Lai, T. L., Ying, Z. L. 1988; 27 (2): 334-358
  • OPEN BANDIT PROCESSES AND OPTIMAL SCHEDULING OF QUEUING-NETWORKS ADVANCES IN APPLIED PROBABILITY Lai, T. L., Ying, Z. L. 1988; 20 (2): 447-472
  • NEARLY OPTIMAL SEQUENTIAL-TESTS OF COMPOSITE HYPOTHESES ANNALS OF STATISTICS Lai, T. L. 1988; 16 (2): 856-886
  • BOUNDARY CROSSING PROBLEMS FOR SAMPLE MEANS ANNALS OF PROBABILITY Lai, T. L. 1988; 16 (1): 375-396
  • HEART-RATE-VARIABILITY DURING RESPIRATORY PAUSES IN PUPPIES AND DOGS PEDIATRIC RESEARCH Haddad, G. G., Jeng, H. J., Lai, T. L. 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 Haddad, G. G., Jeng, H. J., Lai, T. L., Mellins, R. B. 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 Lai, T. L., Wei, C. Z. 1987; 25 (2): 466-481
  • EXTENDED LEAST-SQUARES AND THEIR APPLICATIONS TO ADAPTIVE-CONTROL AND PREDICTION IN LINEAR-SYSTEMS IEEE TRANSACTIONS ON AUTOMATIC CONTROL Lai, T. L., Wei, C. Z. 1986; 31 (10): 898-906
  • WITHIN-BREATH ELECTROMYOGRAPHIC CHANGES DURING LOADED BREATHING IN ADULT SHEEP JOURNAL OF APPLIED PHYSIOLOGY Haddad, G. G., Jeng, H. J., Bazzy, A. R., Lai, T. L. 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 Haddad, G. G., Jeng, H. J., Lai, T. L. 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 Lai, T. L., Robbins, H. 1985; 6 (1): 4-22
  • ORTHONORMAL BANACH SYSTEMS WITH APPLICATIONS TO LINEAR-PROCESSES ZEITSCHRIFT FUR WAHRSCHEINLICHKEITSTHEORIE UND VERWANDTE GEBIETE Lai, T. L., Wei, C. Z. 1985; 70 (3): 381-393
  • OPTIMAL SEQUENTIAL SAMPLING FROM 2 POPULATIONS PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA-PHYSICAL SCIENCES Lai, T. L., Robbins, H. 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

  • RHYTHMIC VARIATIONS IN R-R INTERVAL DURING SLEEP AND WAKEFULNESS IN PUPPIES AND DOGS AMERICAN JOURNAL OF PHYSIOLOGY Haddad, G. G., Jeng, H. J., Lee, S. H., Lai, T. L. 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

  • 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 Lai, T. L., Robbins, H., Yu, K. F. 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 LEISTNER, H. L., Haddad, G. G., Lai, T. L., Mellins, R. B. 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 Haddad, G. G., Gandhi, M. R., Hochwald, G. M., Lai, T. L. 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 Lai, T. L., Siegmund, D. 1983; 11 (2): 478-485
  • CONVERGENCE SYSTEMS AND STRONG CONSISTENCY OF LEAST-SQUARES ESTIMATES IN REGRESSION-MODELS KEXUE TONGBAO Chen, G. J., Lai, T. L., Wei, C. Z. 1983; 28 (1): 16-20
  • ASYMPTOTIC PROPERTIES OF GENERAL AUTOREGRESSIVE MODELS AND STRONG CONSISTENCY OF LEAST-SQUARES ESTIMATES OF THEIR PARAMETERS JOURNAL OF MULTIVARIATE ANALYSIS Lai, T. L., Wei, C. Z. 1983; 13 (1): 1-23
  • LEAST-SQUARES ESTIMATES IN STOCHASTIC REGRESSION-MODELS WITH APPLICATIONS TO IDENTIFICATION AND CONTROL OF DYNAMIC-SYSTEMS ANNALS OF STATISTICS Lai, T. L., Wei, C. Z. 1982; 10 (1): 154-166
  • DETERMINATION OF VENTILATORY PATTERN IN REM-SLEEP IN NORMAL INFANTS JOURNAL OF APPLIED PHYSIOLOGY Haddad, G. G., Lai, T. L., Mellins, R. B. 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 Haddad, G. G., Lai, T. L., EPSTEIN, M. A., EPSTEIN, R. A., Yu, K. F., LEISTNER, H. L., Mellins, R. B. 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 Lai, T. L., Wei, C. Z. 1982; 12 (3): 346-370
  • A LAW OF THE ITERATED LOGARITHM FOR DOUBLE ARRAYS OF INDEPENDENT RANDOM-VARIABLES WITH APPLICATIONS TO REGRESSION AND TIME-SERIES MODELS ANNALS OF PROBABILITY Lai, T. L., Wei, C. Z. 1982; 10 (2): 320-335
  • CONSISTENCY AND ASYMPTOTIC EFFICIENCY OF SLOPE ESTIMATES IN STOCHASTIC-APPROXIMATION SCHEMES ZEITSCHRIFT FUR WAHRSCHEINLICHKEITSTHEORIE UND VERWANDTE GEBIETE Lai, T. L., Robbins, H. 1981; 56 (3): 329-360
  • CONVERGENCE SYSTEMS AND STRONG CONSISTENCY OF LEAST-SQUARES ESTIMATES IN REGRESSION-MODELS JOURNAL OF MULTIVARIATE ANALYSIS GUIJING, C., Lai, T. L., Wei, C. Z. 1981; 11 (3): 319-333
  • VENTILATION AND VENTILATORY PATTERN DURING SLEEP IN ABORTED SUDDEN INFANT DEATH SYNDROME PEDIATRIC RESEARCH Haddad, G. G., LEISTNER, H. L., Lai, T. L., Mellins, R. B. 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 Lai, T. L., Levin, B., Robbins, H., Siegmund, D. 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 LEISTNER, H. L., Haddad, G. G., EPSTEIN, R. A., Lai, T. L., EPSTEIN, M. A., Mellins, R. B. 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 Lai, T. L., Siegmund, D. 1979; 7 (1): 60-76
  • ADAPTIVE DESIGN AND STOCHASTIC-APPROXIMATION ANNALS OF STATISTICS Lai, T. L., Robbins, H. 1979; 7 (6): 1196-1221
  • STRONG CONSISTENCY OF LEAST-SQUARES ESTIMATES IN MULTIPLE-REGRESSION .2. JOURNAL OF MULTIVARIATE ANALYSIS Lai, T. L., Robbins, H., Wei, C. Z. 1979; 9 (3): 343-361
  • LOCAL CONVERGENCE THEOREMS FOR ADAPTIVE STOCHASTIC-APPROXIMATION SCHEMES PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA Lai, T. L., Robbins, H. 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 Lai, T. L., Robbins, H. 1978; 42 (2): 89-111
  • STRONG CONSISTENCY OF LEAST-SQUARES ESTIMATES IN MULTIPLE-REGRESSION PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA Lai, T. L., Robbins, H., Wei, C. Z. 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 Lai, T. L., Robbins, H. 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 Lai, T. L., Robbins, H. 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 Lai, T. L., Siegmund, D. 1977; 5 (5): 946-954
  • STRONG CONSISTENCY OF LEAST-SQUARES ESTIMATES IN REGRESSION-MODELS PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA Lai, T. L., Robbins, H. 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 Lai, T. L., Robbins, H. 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) 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