Significance Tests in Climate Science

被引：27

作者：

Ambaum, Maarten H. P. ^{[1
]}

机构：

[1] Univ Reading, Dept Meteorol, Reading RG6 6BB, Berks, England

来源：

JOURNAL OF CLIMATE | 2010年 / 23卷 / 22期

关键词：

Bayesian methods; Climatology; Statistical techniques;

D O I：

10.1175/2010JCLI3746.1

中图分类号：

P4 [大气科学（气象学）];

学科分类号：

0706 ; 070601 ;

摘要：

A large fraction of papers in the climate literature includes erroneous uses of significance tests. A Bayesian analysis is presented to highlight the meaning of significance tests and why typical misuse occurs. The significance statistic is not a quantitative measure of how confident one can be of the "reality" of a given result. It is concluded that a significance test very rarely provides useful quantitative information.

引用

页码：5927 / 5932

页数：6

共 13 条

[1]

[Anonymous], 1961, The Algebra of Probable Inference

[2]

[Anonymous], 2003, Probability Theory

[3] Significance tests harm progress in forecasting [J].

Armstrong, J. Scott .

INTERNATIONAL JOURNAL OF FORECASTING, 2007, 23 (02) :321-327

[4]

COHEN J, 1994, AM PSYCHOL, V49, P997, DOI 10.1037/0003-066X.50.12.1103

[5] Needed: A ban on the significance test [J].

Hunter, JE .

PSYCHOLOGICAL SCIENCE, 1997, 8 (01) :3-7

[6] PRIOR PROBABILITIES [J].

JAYNES, ET .

IEEE TRANSACTIONS ON SYSTEMS SCIENCE AND CYBERNETICS, 1968, SSC4 (03) :227-&

[7]

Jolliffe I.T., 2004, Weather, V59, P77, DOI [10.1256/wea.132. 03, DOI 10.1256/WEA.132.03]

[8]

Nicholls N, 2001, B AM METEOROL SOC, V82, P981, DOI 10.1175/1520-0477(2001)082<0981:CAATIO>2.3.CO

[9]

[10]

Ryle G., 1949, The concept of mind, DOI 10.4324/9780203875858

← 1 2 →