Prediction, Parsimony and Noise

While many believe the accuracy of scientific models depends on the data used in them, models can actually be more accurate than the data by amplifying hidden patterns or discarding unwanted noise. The use of statistical modeling in research is discussed.

Saved in:
Bibliographic Details
Published in:American scientist 1993-09, Vol.81 (5), p.468-478
Main Author: Gauch, Hugh G.
Format: Article
Language:English
Subjects:
Analysis of variance
Data models
Datasets
Genotypes
Information retrieval noise
Mathematical models
Methods
Modeling
Parsimony
Polynomials
Science
Scientific method
Signal noise
Statistical variance
Statistics
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:While many believe the accuracy of scientific models depends on the data used in them, models can actually be more accurate than the data by amplifying hidden patterns or discarding unwanted noise. The use of statistical modeling in research is discussed.
ISSN:0003-0996
1545-2786