| Abstract: | Acknowledges the corrections and clarifications of the author's (see record 1976-29043-001) equal weights theorem offered by J. E. Laughlin (see record 1979-02690-001) and R. M. Pruzek and B. C. Frederick (see record 1979-02691-001). It is shown that least squares estimators of regression parameters are very unstable under some conditions. Alternative estimation procedures are surveyed, and their widespread utilization is urged. The appropriateness of the conditions of the author's equal weights theorem is discussed. (26 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved) |
