Missing data imputation by utilizing information within incomplete instances

Zhang, Shichao, Jin, Zhi and Zhu, Xiaofeng (2011) Missing data imputation by utilizing information within incomplete instances. Journal of Systems and Software, 84 3: 452-459. doi:10.1016/j.jss.2010.11.887


Author Zhang, Shichao
Jin, Zhi
Zhu, Xiaofeng
Title Missing data imputation by utilizing information within incomplete instances
Journal name Journal of Systems and Software   Check publisher's open access policy
ISSN 0164-1212
1873-1228
Publication date 2011-03-01
Year available 2010
Sub-type Article (original research)
DOI 10.1016/j.jss.2010.11.887
Open Access Status Not yet assessed
Volume 84
Issue 3
Start page 452
End page 459
Total pages 8
Place of publication New York NY, United States
Publisher Elsevier
Language eng
Subject 1712 Software
1710 Information Systems
1708 Hardware and Architecture
Abstract This paper proposes to utilize information within incomplete instances (instances with missing values) when estimating missing values. Accordingly, a simple and efficient nonparametric iterative imputation algorithm, called the NIIA method, is designed for iteratively imputing missing target values. The NIIA method imputes each missing value several times until the algorithm converges. In the first iteration, all the complete instances are used to estimate missing values. The information within incomplete instances is utilized since the second imputation iteration. We conduct some experiments for evaluating the efficiency, and demonstrate: (1) the utilization of information within incomplete instances is of benefit to easily capture the distribution of a dataset; and (2) the NIIA method outperforms the existing methods in accuracy, and this advantage is clearly highlighted when datasets have a high missing ratio.
Keyword Incomplete data analysis
Missing value
Nonparametric imputation
Iterative imputation
Q-Index Code C1
Q-Index Status Confirmed Code
Institutional Status UQ
Additional Notes Available online 30 November 2010

Document type: Journal Article
Sub-type: Article (original research)
Collections: Official 2012 Collection
School of Information Technology and Electrical Engineering Publications
 
Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 35 times in Thomson Reuters Web of Science Article | Citations
Scopus Citation Count Cited 43 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Sun, 20 Mar 2011, 10:08:00 EST