Computational Statistics in Data Science. Группа авторов

       Nola du Toit

      NORC at the University of Chicago

      Chicago, IL

      USA

       Dootika Vats

      Indian Institute of Technology Kanpur

      Kanpur

      India

       Matti Vihola

      University of Jyväskylä

      Jyväskylä

      Finland

       Justin Wang

      University of California at Davis

      Davis, CA

      USA

       Will Wei Sun

      Purdue University

      West Lafayette, IN

      USA

       Leland Wilkinson

      H₂O.ai, Mountain View

      California

      USA

       and

      University of Illinois at Chicago

      Chicago, IL

      USA

       Joong‐Ho Won

      Seoul National University

      Seoul

      South Korea

       Yichao Wu

      University of Illinois at Chicago

      Chicago, IL

      USA

       Min‐ge Xie

      Rutgers University

      Piscataway, NJ

      USA

       Ming Yan

      Michigan State University

      East Lansing, MI

      USA

       Yuling Yao

      Columbia University

      New York, NY

      USA

       and

      Center for Computational Mathematics

      Flatiron Institute

      New York, NY

      USA

       Chun Yip Yau

      Chinese University of Hong Kong

      Shatin

      Hong Kong

       Hao H. Zhang

      University of Arizona

      Tucson, AZ

      USA

       Hua Zhou

      University of California

      Los Angeles, CA

      USA

Preface

      Computational statistics is a core area of modern statistical science and its connections to data science represent an ever‐growing area of study. One of its important features is that the underlying technology changes quite rapidly, riding on the back of advances in computer hardware and statistical software. In this compendium we present a series of expositions that explore the intermediate and advanced concepts, theories, techniques, and practices that act to expand this rapidly evolving field. We hope that scholars and investigators will use the presentations to inform themselves on how modern computational and statistical technologies are applied, and also to build springboards that can develop their further research. Readers will require knowledge of fundamental statistical methods and, depending on the topic of interest they peruse, any advanced statistical aspects necessary to understand and conduct the technical computing procedures.

      The presentation begins with a thoughtful introduction on how we should view Computational Statistics & Data Science in the 21st Century (Holbrook, et al.), followed by a careful tour of contemporary Statistical Software (Schissler, et al.). Topics that follow address a variety of issues, collected into broad topic areas such as Simulation‐based Methods, Statistical Learning, Quantitative Visualization, High‐performance Computing, High‐dimensional Data Analysis, and Numerical Approximations & Optimization.

      Internet access to all of the articles presented here is available via the online collection Wiley StatsRef: Statistics Reference Online (Davidian, et al., 2014–2021); see https://onlinelibrary.wiley.com/doi/book/10.1002/9781118445112.

      From Deep Learning (Li, et al.) to Asynchronous Parallel Computing (Yan), this collection provides a glimpse into how computational statistics may progress in this age of big data and transdisciplinary data science. It is our fervent hope that readers will benefit from it.

      We wish to thank the fine efforts of the Wiley editorial staff, including Kimberly Monroe‐Hill, Paul Sayer, Michael New, Vignesh Lakshmikanthan, Aruna Pragasam, Viktoria Hartl‐Vida, Alison Oliver, and Layla Harden in helping bring this project to fruition.

Tucson, ArizonaSan Diego, California Tucson, ArizonaDavis, California

Walter W. Piegorsch Richard A. Levine Hao Helen Zhang Thomas C. M. Lee

Reference

      1 Davidian, M., Kenett, R.S., Longford, N.T., Molenberghs, G., Piegorsch, W.W., and Ruggeri, F., eds. (2014–2021). Wiley StatsRef: Statistics Reference Online. Chichester: John Wiley & Sons. doi:10.1002/9781118445112.

Part I Computational Statistics and Data Science

1 Computational Statistics and Data Science in the Twenty‐First Century

       Andrew J. Holbrook1, Akihiko Nishimura2, Xiang Ji3, and Marc A. Suchard1

       1University of California, Los Angeles, CA, USA

       2Johns Hopkins University, Baltimore, MD, USA

       3Tulane University, New Orleans, LA, USA

1 Introduction

We are in the midst of the data science revolution. In October 2012, the Harvard Business Review famously declared data scientist the sexiest job of the twenty‐first century [1]. By September 2019, Google searches for the term “data science” had multiplied over sevenfold [2], one multiplicative increase for each intervening year. In the United States between the years 2000 and 2018, the number of bachelor's degrees awarded in either statistics or biostatistics increased over 10‐fold (382–3964), and the number of doctoral degrees almost tripled (249–688) [3]. In 2020, seemingly every major university has established or is establishing its own data science institute, center, or initiative.

Data science [4, 5] combines multiple preexisting disciplines (e.g., statistics, machine learning, and computer science) with a redirected focus on creating,