Statistical computing: Algorithms for computing the sample variance: Analysis and recommendations

Tony F. Chan; Gene H. Golub; Randall J. Leveque

doi:10.1080/00031305.1983.10483115

Statistical computing: Algorithms for computing the sample variance: Analysis and recommendations

Tony F. Chan, Gene H. Golub, Randall J. Leveque

Research output: Contribution to journal › Article › peer-review

208 Scopus citations

Abstract

The problem of computing the variance of a sample of N data points (x), may be difficult for certain data sets, particularly when N is large and the variance is small. We present a survey of possible algorithms and their round-off error bounds, including some new analysis for computations with shifted data. Experimental results confirm these bounds and illustrate the dangers of some algorithms. Specific recommendations are made as to which algorithm should be used in various contexts.

Original language	English (US)
Pages (from-to)	242-247
Number of pages	6
Journal	American Statistician
Volume	37
Issue number	3
DOIs	https://doi.org/10.1080/00031305.1983.10483115
State	Published - Aug 1983
Externally published	Yes

Keywords

Computer algorithms
Round-off errors
Shifted data
Standard deviation
Variance

ASJC Scopus subject areas

Statistics and Probability
General Mathematics
Statistics, Probability and Uncertainty

Access to Document

10.1080/00031305.1983.10483115

Cite this

@article{9c5d97fe6b704169ade40c0e60539e49,

title = "Statistical computing: Algorithms for computing the sample variance: Analysis and recommendations",

abstract = "The problem of computing the variance of a sample of N data points (x), may be difficult for certain data sets, particularly when N is large and the variance is small. We present a survey of possible algorithms and their round-off error bounds, including some new analysis for computations with shifted data. Experimental results confirm these bounds and illustrate the dangers of some algorithms. Specific recommendations are made as to which algorithm should be used in various contexts.",

keywords = "Computer algorithms, Round-off errors, Shifted data, Standard deviation, Variance",

author = "Chan, {Tony F.} and Golub, {Gene H.} and Leveque, {Randall J.}",

year = "1983",

month = aug,

doi = "10.1080/00031305.1983.10483115",

language = "English (US)",

volume = "37",

pages = "242--247",

journal = "American Statistician",

issn = "0003-1305",

publisher = "Taylor and Francis Ltd.",

number = "3",

}

TY - JOUR

T1 - Statistical computing

T2 - Algorithms for computing the sample variance: Analysis and recommendations

AU - Chan, Tony F.

AU - Golub, Gene H.

AU - Leveque, Randall J.

PY - 1983/8

Y1 - 1983/8

N2 - The problem of computing the variance of a sample of N data points (x), may be difficult for certain data sets, particularly when N is large and the variance is small. We present a survey of possible algorithms and their round-off error bounds, including some new analysis for computations with shifted data. Experimental results confirm these bounds and illustrate the dangers of some algorithms. Specific recommendations are made as to which algorithm should be used in various contexts.

AB - The problem of computing the variance of a sample of N data points (x), may be difficult for certain data sets, particularly when N is large and the variance is small. We present a survey of possible algorithms and their round-off error bounds, including some new analysis for computations with shifted data. Experimental results confirm these bounds and illustrate the dangers of some algorithms. Specific recommendations are made as to which algorithm should be used in various contexts.

KW - Computer algorithms

KW - Round-off errors

KW - Shifted data

KW - Standard deviation

KW - Variance

UR - http://www.scopus.com/inward/record.url?scp=84952519631&partnerID=8YFLogxK

U2 - 10.1080/00031305.1983.10483115

DO - 10.1080/00031305.1983.10483115

M3 - Article

AN - SCOPUS:84952519631

SN - 0003-1305

VL - 37

SP - 242

EP - 247

JO - American Statistician

JF - American Statistician

IS - 3

ER -

Statistical computing: Algorithms for computing the sample variance: Analysis and recommendations

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this