NotesFAQContact Us
Search Tips
ERIC Number: ED430994
Record Type: Non-Journal
Publication Date: 1999-Apr
Pages: 11
Abstractor: N/A
Reference Count: N/A
An Illustration of the Least Median Squares (LMS) Regression Using PROGRESS.
Wang, Jianjun
The least mean squares (LS) regression method produced the best linear unbiased estimates under the normal error distribution. However, many researchers have noted that the optimal condition is rarely met in real data analyses. To remedy the impact of potential data contamination, several advantages of the least median squares (LMS) regression are illustrated using a user-friendly software program, "Program for RObust reGRESSion" (PROGRESS). A public data base (the Longitudinal Study of American Youth was carefully chosen to facilitate verification of the empirical comparison between LS and LMS estimation. It is found that the LMS method results in a smaller average error of prediction and covers a larger proportion of variance in regression. In addition, it is demonstrated that even for real data with no significant outliers, the LMS estimator tended to match observations better than the simple LS fit. (Contains 2 tables and 16 references.) (Author/SLD)
Publication Type: Reports - Research; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A