On the eve of the 21st century: Statistical science at a crossroads
暂无分享,去创建一个
Modern statistical science is one of the major scienti c achievements of the 20th century. While many of the concepts had been laid down earlier, it was not until the 20th century that a true science of statistics had come into being. It has been my privilege over the course of my professional life to have known many of the seminal gures in the history of modern 20th century statistics. Regrettably, I did not have a chance to know Karl Pearson and R.A. Fisher, but I did know many of this century’s greatest contributors to statistical theory such as Egon Pearson, Jerzy Neyman, David Blackwell, C.R. Rao, Harold Hotelling, R.C. Bose, Wassily Hoe ding, Gertrude Cox, Jimmie Savage, and Harald Cramer to name a few. Of course, many great contributors are still active today. I often contemplate what these earlier innovators of the 20th century might think about the state of statistical science as we approach the 21st century. The computing revolution we have seen in the last 20 years has thrust upon us great changes in the way we collect data and our general attitude towards both data and methodology. In a real sense, statistical science is at a crossroads. Statistics as a discipline, it seems to me, is all too often de ned by the set of techniques, tools and methodology rather than by the goal of data analysis and inference. De ning the science by its techniques, tools and methods is a prescription for insularity while de ning by its goals is a prescription for openness and expansion. I rather think the great innovators of statistical methodology mentioned above would opt for a rather more goal oriented worldview of statistical science than seems to prevail today in general.
[1] Peter J. Huber,et al. Huge Data Sets , 1994 .
[2] Sanford Weisberg,et al. Computing science and statistics : proceedings of the 30th Symposium on the Interface, Minneapolis, Minnesota, May 13-16, 1998 : dimension reduction, computational complexity and information , 1998 .
[3] Edward J. Wegman,et al. Huge Data Sets and the Frontiers of Computational Feasibility , 1995 .
[4] Peter J. Huber,et al. Issues in Computational Data Analysis , 1992 .