For fitting multiple linear regressions, the ordinary least squares approach is sensitive to outliers and/or violations of model assumptions. The trimmed elemental estimators are more robust to such situations. This package contains functions for computing the trimmed elemental estimates, as well as for creating the bias-corrected and accelerated bootstrap confidence intervals based on elemental regressions.
|Author||Wei Jiang and Matthew S. Mayo|
|Date of publication||2016-06-14 08:29:46|
|Maintainer||Wei Jiang <email@example.com>|
|License||GPL (>= 2)|