
For FullText PDF, please login, if you are a member of IEICE,
or go to Pay Per View on menu list, if you are a nonmember of IEICE.

Influence of Outliers on Estimation Accuracy of Software Development Effort
Kenichi ONO Masateru TSUNODA Akito MONDEN Kenichi MATSUMOTO
Publication
IEICE TRANSACTIONS on Information and Systems
Vol.E104D
No.1
pp.91105 Publication Date: 2021/01/01 Publicized: 2020/10/02 Online ISSN: 17451361
DOI: 10.1587/transinf.2020MPP0005 Type of Manuscript: Special Section PAPER (Special Section on Empirical Software Engineering) Category: Keyword: casebased reasoning, multiple linear regression, effort prediction, outlier,
Full Text: PDF(2.5MB)>>
Summary:
When applying estimation methods, the issue of outliers is inevitable. The extent of their influence has not been clarified, though several studies have evaluated outlier elimination methods. It is unclear whether we should always be sensitive to outliers, whether outliers should always be removed before estimation, and what amount of precaution is required for collecting project data. Therefore, the goal of this study is to illustrate a guideline that suggests how sensitively we should handle outliers. In the analysis, we experimentally add outliers to three datasets, to analyze their influence. We modified the percentage of outliers, their extent (e.g., we varied the actual effort from 100 to 200 personhours when the extent was 100%), the variables including outliers (e.g., adding outliers to function points or effort), and the locations of outliers in a dataset. Next, the effort was estimated using these datasets. We used multiple linear regression analysis and analogy based estimation to estimate the development effort. The experimental results indicate that the influence of outliers on the estimation accuracy is nontrivial when the extent or percentage of outliers is considerable (i.e., 100% and 20%, respectively). In contrast, their influence is negligible when the extent and percentage are small (i.e., 50% and 10%, respectively). Moreover, in some cases, the linear regression analysis was less affected by outliers than analogy based estimation.

