r-lars 1.2 Least angle regression software
Least Angle Regression ("LAR") is a model selection algorithm; a useful and less greedy version of traditional forward selection methods. A simple modification of the LAR algorithm implements Tibshirani's Lasso; the Lasso modification of LARS calculates the entire Lasso path of coefficients for a given problem at the cost of a single least squares fit. Another LARS modification efficiently implements epsilon Forward Stagewise linear regression.
- Website: https://web.stanford.edu/~hastie/Papers/LARS/
- License: GPL 2
- Package source: statistics.scm
- Patches: None
- Builds: x86_64-linux, i686-linux