Inherent Complexity: A Problem for Statistical Model Evaluation

被引：3

作者：

Romeijn, Jan-Willem ^{[1
]}

机构：

[1] Univ Groningen, Oude Boteringestr 52, NL-9712 GL Groningen, Netherlands

来源：

PHILOSOPHY OF SCIENCE | 2017年 / 84卷 / 05期

关键词：

SELECTION;

D O I：

10.1086/693927

中图分类号：

N09 [自然科学史]; B [哲学、宗教];

学科分类号：

01 ; 0101 ; 010108 ; 060207 ; 060305 ; 0712 ;

摘要：

This article investigates a problem for statistical model evaluation, in particular for curve fitting: by employing a different family of curves we can fit any scatter plot almost perfectly at apparently minor cost in terms of model complexity. The problem is resolved by an appeal to prior probabilities. This leads to some general lessons about how to approach model evaluation.

引用

页码：797 / 809

页数：13

共 14 条

[1] Akaike H., 1992, 2 INT S INF THEOR, P610, DOI [10.1007/978-1-4612-1694-0, 10.1007/978-1-4612-0919-538, 10.1007/978-1-4612-0919-5_38, 10.1007/978-0-387-98135-2, DOI 10.1007/978-1-4612-0919-538]
[2] [Anonymous], 2002, Model selection and multimodel inference: a practical informationtheoretic approach
[3] [Anonymous], 2006, Model selection and model averaging, DOI DOI 10.1017/CBO9780511790485.003
[4] Balasubramanian V, 2005, NEU INF PRO, P81
[5] MODEL SELECTION AND AKAIKE INFORMATION CRITERION (AIC) - THE GENERAL-THEORY AND ITS ANALYTICAL EXTENSIONS
BOZDOGAN, H
[J]. PSYCHOMETRIKA, 1987, 52 (03) : 345 - 370
[6] Grnwald P. D., 2007, The Minimum Description Length Principle
[7] Harman G., 2007, Reliable Reasoning
[8] Counting probability distributions: Differential geometry and model selection
Myung, IJ
Balasubramanian, V
Pitt, MA
[J]. PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 2000, 97 (21) : 11170 - 11175
[9] Bayesian model selection in social research
Raftery, AE
[J]. SOCIOLOGICAL METHODOLOGY 1995, VOL 25, 1995, 25 : 111 - 163
[10] Romeijn J.W., 2008, BAYESIAN EVALUATION

← 1 2 →