Classical and a robust method in dealing with outlier in factorial designs : an empirical example.
Valid analysis of an experimental design data requires several assumptions, such as normality, constant variances and independency. In practice, those assumptions can be violated due to causes, such as the presence of an outlying observation. A more appropriate and modern approach is needed, that...
| Main Authors: | , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English English |
| Published: |
2012
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/27587/1/ID%2027587.pdf |
| Summary: | Valid analysis of an experimental design data requires several assumptions, such as normality, constant
variances and independency. In practice, those assumptions can be violated due to causes, such as the presence
of an outlying observation. A more appropriate and modern approach is needed, that is to use a robust procedure that provides estimation, inference and testing that are not influenced by outlying observations but describes correctly the structure for the bulk of the data. A well-known approach to handle dataset with outliers is the M-estimation. In this paper, both classical and robust procedures are employed to data of a factorial experiment. And we point out that relying on classical method instead of robust methods lead to misleading conclusion of the analysis. |
|---|