Past, Present, and Future of Statistical Science

Table of contents

1. Cover
2. Half Title
3. Title Page
4. Copyright Page
5. Contents (1/3)
6. Contents (2/3)
7. Contents (3/3)
8. Preface
9. Contributors
10. I. The history of COPSS
    1. 1. A brief history of the Committee of Presidents of Statistical Societies (COPSS)
       1. 1.1. Introduction
       2. 1.2. COPSS activities in the early years
       3. 1.3. COPSS activities in recent times
       4. 1.4. Awards (1/3)
       5. 1.4. Awards (2/3)
       6. 1.4. Awards (3/3)
11. II. Reminiscences and personal reflections on career paths
    1. 2. Reminiscences of the Columbia University Department of Mathematical Statistics in the late 1940s
       1. 2.1. Introduction: Pre-Columbia
       2. 2.2. Columbia days
       3. 2.3. Courses
    2. 3. A career in statistics
       1. 3.1. Education
       2. 3.2. Postdoc at University of Chicago
       3. 3.3. University of Illinois and Stanford
       4. 3.4. MIT and Harvard
    3. 4. “. . . how wonderful the field of statistics is. . . ”
       1. 4.1. Introduction
       2. 4.2. The speech (edited some)
       3. 4.3. Conclusion
    4. 5. An unorthodox journey to statistics: Equity issues, remarks on multiplicity
       1. 5.1. Pre-statistical career choices
       2. 5.2. Becoming a statistician
       3. 5.3. Introduction to and work in multiplicity
       4. 5.4. General comments on multiplicity
    5. 6. Statistics before and after my COPSS Prize
       1. 6.1. Introduction
       2. 6.2. The foundation of mathematical statistics
       3. 6.3. My work before 1979
       4. 6.4. My work after 1979
       5. 6.5. Some observations (1/2)
       6. 6.5. Some observations (2/2)
    6. 7. The accidental biostatistics professor
       1. 7.1. Public school and passion for mathematics
       2. 7.2. College years and discovery of statistics
       3. 7.3. Thwarted employment search after college
       4. 7.4. Graduate school as a fallback option
       5. 7.5. Master’s degree in statistics at Purdue
       6. 7.6. Thwarted employment search after Master’s degree
       7. 7.7. Graduate school again as a fallback option
       8. 7.8. Dissertation research and family issues
       9. 7.9. Job offers — finally!
       10. 7.10. Four years at UNC-Chapel Hill
       11. 7.11. Thirty-three years at Emory University
       12. 7.12. Summing up and acknowledgements
    7. 8. Developing a passion for statistics
       1. 8.1. Introduction
       2. 8.2. The first statistical seeds
       3. 8.3. Graduate training
       4. 8.4. The PhD
       5. 8.5. Job and postdoc hunting
       6. 8.6. The postdoc years
       7. 8.7. Starting on the tenure track
    8. 9. Reflections on a statistical career and their implications
       1. 9.1. Early years
       2. 9.2. Statistical diagnostics
       3. 9.3. Optimal experimental design
       4. 9.4. Enjoying statistical practice
       5. 9.5. A lesson learned
    9. 10. Science mixes it up with statistics
       1. 10.1. Introduction
       2. 10.2. Collaborators
       3. 10.3. Some collaborative projects
       4. 10.4. Conclusions
    10. 11. Lessons from a twisted career path
        1. 11.1. Introduction
        2. 11.2. Student days
        3. 11.3. Becoming a researcher
        4. 11.4. Final thoughts
    11. 12. Promoting equity
        1. 12.1. Introduction
        2. 12.2. The Elizabeth Scott Award
        3. 12.3. Insurance
        4. 12.4. Title IX
        5. 12.5. Human rights
        6. 12.6. Underrepresented groups
12. III. Perspectives on the field and profession
    1. 13. Statistics in service to the nation
       1. 13.1. Introduction
       2. 13.2. The National Halothane Study
       3. 13.3. The President’s Commission and CNSTAT
       4. 13.4. Census-taking and multiple-systems estimation
       5. 13.5. Cognitive aspects of survey methodology
       6. 13.6. Privacy and confidentiality
       7. 13.7. The accuracy of the polygraph
       8. 13.8. Take-home messages
    2. 14. Where are the majors?
       1. 14.1. The puzzle
       2. 14.2. The data
       3. 14.3. Some remarks
    3. 15. We live in exciting times
       1. 15.1. Introduction
       2. 15.2. Living with change
       3. 15.3. Living the revolution (1/2)
       4. 15.3. Living the revolution (2/2)
    4. 16. The bright future of applied statistics
       1. 16.1. Introduction
       2. 16.2. Becoming an applied statistician
       3. 16.3. Genomics and the measurement revolution
       4. 16.4. The bright future
    5. 17. The road travelled: From statistician to statistical scientist
       1. 17.1. Introduction
       2. 17.2. Kin-cohort study: My gateway to genetics
       3. 17.3. Gene-environment interaction: Bridging
       4. 17.4. Genome-wide association studies (GWAS):
       5. 17.5. The post-GWAS era: What does it all mean?
       6. 17.6. Conclusion
    6. 18. A journey into statistical genetics and genomics
       1. 18.1. The ’omics era
       2. 18.2. My move into statistical genetics and genomics
       3. 18.3. A few lessons learned
       4. 18.4. A few emerging areas in statistical genetics and
       5. 18.5. Training the next generation statistical genetic and
       6. 18.6. Concluding remarks
    7. 19. Reflections on women in statistics in Canada
       1. 19.1. A glimpse of the hidden past
       2. 19.2. Early historical context
       3. 19.3. A collection of firsts for women
       4. 19.4. Awards
       5. 19.5. Builders
       6. 19.6. Statistical practice
       7. 19.7. The current scene
    8. 20. “The whole women thing”
       1. 20.1. Introduction
       2. 20.2. “How many women are there in your department?”
       3. 20.3. “Should I ask for more money?”
       4. 20.4. “I’m honored”
       5. 20.5. “I loved that photo”
       6. 20.6. Conclusion
    9. 21. Reflections on diversity
       1. 21.1. Introduction
       2. 21.2. Initiatives for minority students
       3. 21.3. Impact of the diversity programs
       4. 21.4. Gender issues
13. IV. Reflections on the discipline
    1. 22. Why does statistics have two theories?
       1. 22.1. Introduction
       2. 22.2. 65 years and what’s new
       3. 22.3. Where do the probabilities come from?
       4. 22.4. Inference for regular models: Frequency
       5. 22.5. Inference for regular models: Bootstrap
       6. 22.6. Inference for regular models: Bayes
       7. 22.7. The frequency-Bayes contradiction
       8. 22.8. Discussion
    2. 23. Conditioning is the issue
       1. 23.1. Introduction
       2. 23.2. Cox example and a pedagogical example
       3. 23.3. Likelihood and stopping rule principles
       4. 23.4. What it means to be a frequentist
       5. 23.5. Conditional frequentist inference
       6. 23.6. Final comments
    3. 24. Statistical inference from a Dempster–Shafer perspective
       1. 24.1. Introduction
       2. 24.2. Personal probability
       3. 24.3. Personal probabilities of “don’t know”
       4. 24.4. The standard DS protocol
       5. 24.5. Nonparametric inference
       6. 24.6. Open areas for research
    4. 25. Nonparametric Bayes
       1. 25.1. Introduction
       2. 25.2. A brief history of NP Bayes
       3. 25.3. Gazing into the future (1/2)
       4. 25.3. Gazing into the future (2/2)
    5. 26. How do we choose our default methods?
       1. 26.1. Statistics: The science of defaults
       2. 26.2. Ways of knowing
       3. 26.3. The pluralist’s dilemma
       4. 26.4. Conclusions
    6. 27. Serial correlation and Durbin–Watson bounds
       1. 27.1. Introduction
       2. 27.2. Circular serial correlation
       3. 27.3. Periodic trends
       4. 27.4. Uniformly most powerful tests
       5. 27.5. Durbin–Watson
    7. 28. A non-asymptotic walk in probability and statistics
       1. 28.1. Introduction
       2. 28.2. Model selection
       3. 28.3. Welcome to Talagrand’s wonderland
       4. 28.4. Beyond Talagrand’s inequality
    8. 29. The past’s future is now: What will the present’s future bring?
       1. 29.1. Introduction
       2. 29.2. Symbolic data
       3. 29.3. Illustrations (1/2)
       4. 29.3. Illustrations (2/2)
       5. 29.4. Conclusion
    9. 30. Lessons in biostatistics
       1. 30.1. Introduction
       2. 30.2. It’s the science that counts
       3. 30.3. Immortal time
       4. 30.4. Multiplicity
       5. 30.5. Conclusion
    10. 31. A vignette of discovery
        1. 31.1. Introduction
        2. 31.2. CMV infection and clinical pneumonia
        3. 31.3. Interventions
        4. 31.4. Conclusions
    11. 32. Statistics and public health research
        1. 32.1. Introduction
        2. 32.2. Public health research
        3. 32.3. Biomarkers and nutritional epidemiology
        4. 32.4. Preventive intervention development and testing
        5. 32.5. Clinical trial data analysis methods
        6. 32.6. Summary and conclusion
    12. 33. Statistics in a new era for finance and health care
        1. 33.1. Introduction
        2. 33.2. Comparative effectiveness research clinical studies
        3. 33.3. Innovative clinical trial designs in translational
        4. 33.4. Credit portfolios and dynamic empirical Bayes in
        5. 33.5. Statistics in the new era of finance
        6. 33.6. Conclusion
    13. 34. Meta-analyses: Heterogeneity can be a good thing
        1. 34.1. Introduction
        2. 34.2. Early years of random effects for meta-analysis
        3. 34.3. Random effects and clinical trials
        4. 34.4. Meta-analysis in genetic epidemiology
        5. 34.5. Conclusions
    14. 35. Good health: Statistical challenges in personalizing disease prevention
        1. 35.1. Introduction
        2. 35.2. How do we personalize disease risks?
        3. 35.3. How do we evaluate a personal risk model?
        4. 35.4. How do we estimate model performance measures?
        5. 35.5. Can we improve how we use epidemiological data
        6. 35.6. Concluding remarks
    15. 36. Buried treasures
        1. 36.1. Three short stories
        2. 36.2. Concluding remarks
    16. 37. Survey sampling: Past controversies, current orthodoxy, and future paradigms
        1. 37.1. Introduction
        2. 37.2. Probability or purposive sampling?
        3. 37.3. Design-based or model-based inference? (1/2)
        4. 37.3. Design-based or model-based inference? (2/2)
        5. 37.4. A unified framework: Calibrated Bayes
        6. 37.5. Conclusions
    17. 38. Environmental informatics: Uncertainty quantification in the environmental sciences
        1. 38.1. Introduction
        2. 38.2. Hierarchical statistical modeling
        3. 38.3. Decision-making in the presence of uncertainty
        4. 38.4. Smoothing the data
        5. 38.5. EI for spatio-temporal data (1/2)
        6. 38.5. EI for spatio-temporal data (2/2)
        7. 38.6. The knowledge pyramid
        8. 38.7. Conclusions
    18. 39. A journey with statistical genetics
        1. 39.1. Introduction
        2. 39.2. The 1970s: Likelihood inference and the
        3. 39.3. The 1980s: Genetic maps and hidden Markov
        4. 39.4. The 1990s: MCMC and complex stochastic systems
        5. 39.5. The 2000s: Association studies and gene expression
        6. 39.6. The 2010s: From association to relatedness
        7. 39.7. To the future
    19. 40. Targeted learning: From MLE to TMLE
        1. 40.1. Introduction
        2. 40.2. The statistical estimation problem
        3. 40.3. The curse of dimensionality for the MLE
        4. 40.4. Super learning
        5. 40.5. Targeted learning
        6. 40.6. Some special topics
        7. 40.7. Concluding remarks
    20. 41. Statistical model building, machine learning, and the ah-ha moment
        1. 41.1. Introduction: Manny Parzen and RKHS
        2. 41.2. Regularization methods, RKHS and sparse models
        3. 41.3. Remarks on the nature-nurture debate,
        4. 41.4. Conclusion
    21. 42. In praise of sparsity and convexity
        1. 42.1. Introduction
        2. 42.2. Sparsity, convexity and
        3. 42.3. An example
        4. 42.4. The covariance test
        5. 42.5. Conclusion
    22. 43. Features of Big Data and sparsest solution in high confidence set
        1. 43.1. Introduction
        2. 43.2. Heterogeneity
        3. 43.3. Computation
        4. 43.4. Spurious correlation
        5. 43.5. Incidental endogeneity
        6. 43.6. Noise accumulation
        7. 43.7. Sparsest solution in high confidence set
        8. 43.8. Conclusion
    23. 44. Rise of the machines
        1. 44.1. Introduction
        2. 44.2. The conference culture
        3. 44.3. Neglected research areas
        4. 44.4. Case studies
        5. 44.5. Computational thinking
        6. 44.6. The evolving meaning of data
        7. 44.7. Education and hiring
        8. 44.8. If you can’t beat them, join them
    24. 45. A trio of inference problems that could win you a Nobel Prize in statistics (if you help fund it)
        1. 45.1. Nobel Prize? Why not COPSS?
        2. 45.2. Multi-resolution inference (1/2)
        3. 45.2. Multi-resolution inference (2/2)
        4. 45.3. Multi-phase inference (1/2)
        5. 45.3. Multi-phase inference (2/2)
        6. 45.4. Multi-source inference (1/2)
        7. 45.4. Multi-source inference (2/2)
        8. 45.5. The ultimate prize or price (1/2)
        9. 45.5. The ultimate prize or price (2/2)
14. V. Advice for the next generation
    1. 46. Inspiration, aspiration, ambition
       1. 46.1. Searching the source of motivation
       2. 46.2. Examples of inspiration, aspiration, and ambition
       3. 46.3. Looking to the future
    2. 47. Personal reflections on the COPSS Presidents’ Award
       1. 47.1. The facts of the award
       2. 47.2. Persistence
       3. 47.3. Luck: Have a wonderful Associate Editor
       4. 47.4. Find brilliant colleagues
       5. 47.5. Serendipity with data
       6. 47.6. Get fascinated: Heteroscedasticity
       7. 47.7. Find smart subject-matter collaborators
       8. 47.8. After the Presidents’ Award
    3. 48. Publishing without perishing and other career advice
       1. 48.1. Introduction
       2. 48.2. Achieving balance, and how you never know
       3. 48.3. Write it, and write it again
       4. 48.4. Parting thoughts
    4. 49. Converting rejections into positive stimuli
       1. 49.1. My first attempt
       2. 49.2. I’m learning
       3. 49.3. My first JASA submission
       4. 49.4. Get it published!
       5. 49.5. Find reviewers who understand
       6. 49.6. Sometimes it’s easy, even with errors
       7. 49.7. It sometimes pays to withdraw the paper!
       8. 49.8. Conclusion
    5. 50. The importance of mentors
       1. 50.1. My early years
       2. 50.2. The years at Princeton University
       3. 50.3. Harvard University — the early years
       4. 50.4. My years in statistics as a PhD student
       5. 50.5. The decade at ETS
       6. 50.6. Interim time in DC at EPA, at the University of
       7. 50.7. The three decades at Harvard
       8. 50.8. Conclusions
    6. 51. Never ask for or give advice, make mistakes, accept mediocrity, enthuse
       1. 51.1. Never ask for or give advice
       2. 51.2. Make mistakes
       3. 51.3. Accept mediocrity
       4. 51.4. Enthuse
    7. 52. Thirteen rules
       1. 52.1. Introduction
       2. 52.2. Thirteen rules for giving a really bad talk