The effect of log transformation in multivariate liner regression models for software effort prediction

Akito Monden, Kenichi Kobayashi

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Multivariate linear regression models have been commonly used as software efort prediction models. To improve the prediction accuracy, it is a common practice to transform (especially, log-transform) the data before building a model, although its theoretical basis is not necessarily clear. This paper reveals that the log-transformed linear regression model (log-log regression model) is equal to the exponential model, which is suitable to characterize various relationships among software related metrics. However, when using a log-log regression model, the result of inverse transformation tends to under-estimate the efort. This paper also introduces a method to correct such bias.

Original languageEnglish
Pages (from-to)234-239
Number of pages6
JournalComputer Software
Volume27
Issue number4
Publication statusPublished - 2010
Externally publishedYes

Fingerprint

Linear regression
Mathematical transformations

ASJC Scopus subject areas

  • Software

Cite this

The effect of log transformation in multivariate liner regression models for software effort prediction. / Monden, Akito; Kobayashi, Kenichi.

In: Computer Software, Vol. 27, No. 4, 2010, p. 234-239.

Research output: Contribution to journalArticle

@article{7a7274866bab44cd886301bfb800f27e,
title = "The effect of log transformation in multivariate liner regression models for software effort prediction",
abstract = "Multivariate linear regression models have been commonly used as software efort prediction models. To improve the prediction accuracy, it is a common practice to transform (especially, log-transform) the data before building a model, although its theoretical basis is not necessarily clear. This paper reveals that the log-transformed linear regression model (log-log regression model) is equal to the exponential model, which is suitable to characterize various relationships among software related metrics. However, when using a log-log regression model, the result of inverse transformation tends to under-estimate the efort. This paper also introduces a method to correct such bias.",
author = "Akito Monden and Kenichi Kobayashi",
year = "2010",
language = "English",
volume = "27",
pages = "234--239",
journal = "Computer Software",
issn = "0289-6540",
publisher = "Japan Society for Software Science and Technology",
number = "4",

}

TY - JOUR

T1 - The effect of log transformation in multivariate liner regression models for software effort prediction

AU - Monden, Akito

AU - Kobayashi, Kenichi

PY - 2010

Y1 - 2010

N2 - Multivariate linear regression models have been commonly used as software efort prediction models. To improve the prediction accuracy, it is a common practice to transform (especially, log-transform) the data before building a model, although its theoretical basis is not necessarily clear. This paper reveals that the log-transformed linear regression model (log-log regression model) is equal to the exponential model, which is suitable to characterize various relationships among software related metrics. However, when using a log-log regression model, the result of inverse transformation tends to under-estimate the efort. This paper also introduces a method to correct such bias.

AB - Multivariate linear regression models have been commonly used as software efort prediction models. To improve the prediction accuracy, it is a common practice to transform (especially, log-transform) the data before building a model, although its theoretical basis is not necessarily clear. This paper reveals that the log-transformed linear regression model (log-log regression model) is equal to the exponential model, which is suitable to characterize various relationships among software related metrics. However, when using a log-log regression model, the result of inverse transformation tends to under-estimate the efort. This paper also introduces a method to correct such bias.

UR - http://www.scopus.com/inward/record.url?scp=78650819653&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78650819653&partnerID=8YFLogxK

M3 - Article

VL - 27

SP - 234

EP - 239

JO - Computer Software

JF - Computer Software

SN - 0289-6540

IS - 4

ER -