Statistics Overview - Concepedia

Principles

Definition

Statistics is the academic discipline and methodological approach concerned with the systematic collection, organization, analysis, interpretation, and presentation of quantitative data. It provides principles and techniques for understanding variability, making inferences about populations based on sample evidence, and informing decision-making under uncertainty.

Ontological type

Core Methods

Theoretical Foundations

Uncertainty and Inference

Key developments

Key Figures

Foundations of Statistical Theory era

T. W. Anderson [1] was a central figure in this era, with affiliations including Columbia University [3] and the University of Cambridge [4]. Anderson [1]'s 1952 paper Asymptotic Theory of Certain "Goodness of Fit" Criteria Based on Stochastic Processes [7] advanced the asymptotic understanding of goodness-of-fit criteria and their role in evaluating model adequacy, a cornerstone for rigorous inference in this era. Ronald Aylmer Fisher [2] was a foundational figure in this era, with affiliations at Harvard University [5] and Cornell University [6]. Fisher [2]'s 1925 Theory of Statistical Estimation [8] laid the groundwork for modern estimation theory, introducing key ideas such as estimator efficiency and the role of information in inference, which profoundly shaped statistical practice in this era.

Computation and Model Selection era

Keith E. Muller [1] was a leading statistician whose collaborative work spanned Harvard University [3] and the University of California, San Diego [4] during the Computation and Model Selection era. His contributions to this era include methodological foundations for power analysis as illustrated by Statistical Power Analysis for the Behavioral Sciences [7], which provided practical guidelines for estimating sample size and statistical power and thereby improved study design under growing computational capacity. Hirotugu Akaike [2] was a pivotal figure affiliated with Texas A&M University [5] and the Indian Statistical Institute [6] during this period. His 1974 work A new look at the statistical model identification [8] introduced model identification principles and the information-criterion framework that shaped model selection and regularization approaches in high-dimensional settings.

Hierarchical Reproducible Modeling era

David Moher [1] is a prominent figure associated with Stanford University [3] and Johns Hopkins University [4]. His key contributions include shaping the PRISMA statement [6] and PRISMA Statement [7], standardized reporting guidelines that advanced transparent, reproducible evidence synthesis in hierarchical modeling. Douglas G. Altman [2] is associated with the Institute of Cancer Research [5] and Stanford University [3] in this era. His work on the PRISMA statements [6][7] helped establish rigorous reporting standards that underpin reproducible meta-analytic workflows in hierarchical models.