Missing values
From Wikipedia, the free encyclopedia
In statistics, missing values are a common occurrence. Several statistical methods have been developed to deal with this problem. Missing values mean that no data value is stored for the variable in the current observation. Modern statistical packages made dealing with missing values much easier.
[edit] Techniques of dealing with missing values
- Imputation (statistics)
- Single imputation
- Multiple imputatuion
- EM imputation (also known as expectation-maximization imputation, see Expectation-maximization algorithm)
- full information maximum likelihood estimation
- indicator variable
- Listwise deletion/casewise deletion
- Pairwise deletion
- Mean substitution
- Mplus
- MCAR (missing completely at random)
[edit] Further reading
- Acock, A. C, Working With Missing Values, JOURNAL OF MARRIAGE AND FAMILY, 2005, VOL 67; NUMBER 4, pages 1012-1028
- Jan Van den Broeck, Solveig Argeseanu Cunningham, Roger Eeckels, and Kobus Herbst, Data Cleaning: Detecting, Diagnosing, and Editing Data Abnormalities, PLoS Med. 2005 October; 2(10): e267. [1]