Lasso2 (L1 Constrained Estimation Routines)

Just a copy of http://www.maths.uwa.edu.au/~berwin/software/lasso.html as of 2002-03-11.

Release 2.1, released on 9 December 1999

Via this page you may access two shar files with C routines for solving regression problems in which a constraint is imposed on the sum of the absolute values of the parameter estimates. The shar file also includes interface routines to S-PLUS and documentation files. To see what changed since the last release, check the ChangeLog.

The first shar file contains just the C and S-PLUS code and a manual for the library (roughly 950kB).
The second shar file contains additionally postscript files for most of the manuscripts mentioned below (roughly 2.5MB).

The S-PLUS interface was developed together with Bill Venables and uses the modelling functions of S-PLUS. The underlying C engine is based on the algorithm described in

Osborne, M.R., Presnell, B. and Turlach, B.A. (1999). On the LASSO and its dual, Journal of Computational and Graphical Statistics. To appear.

Justin Lokhorst extended during his honours project the library to include generalised linear models.
The work was motivated by the following article

Tibshirani, R. (1996). Regression shrinkage and selection via the lasso, Journal of the Royal Statistical Society, Series B 58(1): 267-288.

Further discussion and applications of this methodology can be found in

Osborne, M.R. (1998). Variable selection and control in least squares problems, Mathematics Research Report MRR98-047, Centre for Mathematics and its Applications, School of Mathematical Sciences, Australian National University, Canberra ACT 0200.

Osborne, M.R., Presnell, B. and Turlach, B.A. (1998). Knot selection for regression splines via the lasso, in S. Weisberg (ed.), Dimension Reduction, Computational Complexity, and Information, Vol. 30 of Computing Science and Statistics, Interface Foundation of North America, Inc., Fairfax Station, VA 22039-7460, pp. 44-49.

Osborne, M.R., Presnell, B. and Turlach, B.A. (2000). A new approach to variable selection in least squares problems, IMA Journal of Numerical Analysis. To appear.

This library is distributed in the hope that it may also be useful to others. Hence, this software is provided "as is" without any expressed or implied warranty.

The smaller shar file is also available in the S Archive on StatLib. (Note that this may be an older version)

Author: Berwin A Turlach
Date Last modified:Mon Mar 11 16:45:27 WST 2002
Feedback: please direct comments about this page to berwin@maths.uwa.edu.au
URL: http://www.maths.uwa.edu.au/~berwin/software/lasso.html