Jason Liao
2005-Jun-08 18:42 UTC
[Rd] feature request: enhance solve function with a symmetric option for faster operation for symmetric matrix
Dear R developers,
Recently I have a simulation program that needs to solve a large number
of linear regression A x = b, where A is symmetric and constantly
chnages. I found that the solve function is considerably less efficient
than
solve_symmetric = function(A, B)
{
chol2inv(chol(A)) %*% B;
}
So I would request that the solve function be enhanced with an
symmetric option, which defaults to FALSE, the current solve. But when
symmetric=TRUE, a more efficient algorithm for symmetric matrix is
used. Thanks.
Jason
Jason Liao, http://www.geocities.com/jg_liao
Dept. of Biostatistics, http://www2.umdnj.edu/bmtrxweb
University of Medicine and Dentistry of New Jersey
683 Hoes Lane West, Piscataway? NJ 08854
phone 732-235-5429, School of Public Health office
phone 732-235-9824, Cancer Institute of New Jersey office
Douglas Bates
2005-Jun-08 19:07 UTC
[Rd] feature request: enhance solve function with a symmetric option for faster operation for symmetric matrix
On 6/8/05, Jason Liao <jg_liao@yahoo.com> wrote:> Dear R developers, > > Recently I have a simulation program that needs to solve a large number > of linear regression A x = b, where A is symmetric and constantly > chnages. I found that the solve function is considerably less efficient > than > > solve_symmetric = function(A, B) > { > chol2inv(chol(A)) %*% B; > } > > So I would request that the solve function be enhanced with an > symmetric option, which defaults to FALSE, the current solve. But when > symmetric=TRUE, a more efficient algorithm for symmetric matrix is > used. Thanks. > > Jason > > > Jason Liao, http://www.geocities.com/jg_liao > Dept. of Biostatistics, http://www2.umdnj.edu/bmtrxweb > University of Medicine and Dentistry of New Jersey > 683 Hoes Lane West, Piscataway? NJ 08854 > phone 732-235-5429, School of Public Health office > phone 732-235-9824, Cancer Institute of New Jersey office > > ______________________________________________ > R-devel@stat.math.ethz.ch mailing list > https://stat.ethz.ch/mailman/listinfo/r-devel >Try the dsyMatrix or dpoMatrix class in the Matrix package.