Software effort prediction models using maximum likelihood methods require multivariate normality of the software metrics data sample: Can such a sample be made multivariate normal?

研究成果: Conference article同行評審

1 引文 斯高帕斯(Scopus)

摘要

Missing data often appear in software metrics data samples used to construct software effort prediction models1. So far, the least biased and thus the most strongly recommended family of such models capable of handling missing data are those using maximum likelihood methods. However, the theory of such maximum likelihood methods assumes that the data samples underlying the model construction are multivariate normal. Previous researches on such models simply ignored the violation of such an assumption by the empirical data samples. This paper proposes and empirically illustrates a not-so-complicated but effective technique to transform the data sample for the purpose of meeting such an assumption. This technique is empirically proven to work for typical software metrics data samples and the author recommends applying such a technique in any further researches on and practical industrial application of software effort prediction models using maximum likelihood methods.

原文English
頁(從 - 到)274-279
頁數6
期刊Proceedings - International Computer Software and Applications Conference
1
出版狀態Published - 2004
事件Proceedings of the 28th Annual International Computer Software and Applications Conference, COMPSAC 2004 - Hong Kong, China, Hong Kong
持續時間: 28 9月 200430 9月 2004

指紋

深入研究「Software effort prediction models using maximum likelihood methods require multivariate normality of the software metrics data sample: Can such a sample be made multivariate normal?」主題。共同形成了獨特的指紋。

引用此