Online Public Access Catalogue (OPAC)
Library,Documentation and Information Science Division

“A research journal serves that narrow

borderland which separates the known from the unknown”

-P.C.Mahalanobis


Image from Google Jackets

Seven pillars of statistical wisdom / Stephen M. Stigler.

By: Material type: TextTextPublication details: Cambridge : Harvard University Press, ©2016.Description: 230 pages : illustrations ; 18 cmISBN:
  • 9780674088917 (pbk. : alk. paper)
Subject(s): DDC classification:
  • 000SA.01 23 St855
Contents:
1. Aggregation: from tables and means to least squares -- 2. Information: its measurement and rate of change -- 3. Likelihood: calibration on a probability scale -- 4. Intercomparison: within-sample variation as a standard -- 5. Regression: multivariate analysis, Bayesian inference, and causal inference -- 6. Design: experimental planning and the role of randomization -- 7. Residual: scientific logic, model comparison, and diagnostic display.
Summary: A summary of the seven most consequential ideas in the history of statistics, ideas that have proven their importance over a century or more and yet still define the basis of statistical science in the present day. Separately each was a radical idea when introduced, and most remain radical today when they are extended to new territory. Together they define statistics as a scientific field in a way that differentiates it from mathematics and computer science, fields which partner with statistics today but also maintain their separate identities. These "pillars" are presented in their historical context, and some flavor of their development and variety of forms is also given in historical context. The framework of these seven is quite different from the usual ways statistical ideas are arranged, such as in most courses on the subject, and thus they give a new way to think about statistics.
Tags from this library: No tags from this library for this title. Log in to add tags.

Includes bibliographical references and index.

1. Aggregation: from tables and means to least squares --
2. Information: its measurement and rate of change --
3. Likelihood: calibration on a probability scale --
4. Intercomparison: within-sample variation as a standard --
5. Regression: multivariate analysis, Bayesian inference, and causal inference --
6. Design: experimental planning and the role of randomization -- 7. Residual: scientific logic, model comparison, and diagnostic display.

A summary of the seven most consequential ideas in the history of statistics, ideas that have proven their importance over a century or more and yet still define the basis of statistical science in the present day. Separately each was a radical idea when introduced, and most remain radical today when they are extended to new territory. Together they define statistics as a scientific field in a way that differentiates it from mathematics and computer science, fields which partner with statistics today but also maintain their separate identities. These "pillars" are presented in their historical context, and some flavor of their development and variety of forms is also given in historical context. The framework of these seven is quite different from the usual ways statistical ideas are arranged, such as in most courses on the subject, and thus they give a new way to think about statistics.

There are no comments on this title.

to post a comment.
Library, Documentation and Information Science Division, Indian Statistical Institute, 203 B T Road, Kolkata 700108, INDIA
Phone no. 91-33-2575 2100, Fax no. 91-33-2578 1412, ksatpathy@isical.ac.in