README.md

In jimb3/QREcon: Economical QR decomposition

QREcon

The goal of QREcon is to be simple alternative to doing QR decomposition in R that is intuitive and easy to use.

Installation

You can install the released version of QREcon from github with:

devtools::install_github("jimb3/QREcon")

Example

This is a example showing how to use QREcon returning a list:

## example returning a list with Q and R
library(QREcon)
# matrix to perform QR decomposition on
x <- matrix(c(1.,1.,1.,2.,3.,5.), 3, 2)
# call to QR decomposition routine
qr <- QREcon(x)
# view result, if q and r are NUll decomposition failed
qr
#> $q
#>            [,1]       [,2]
#> [1,] -0.5773503  0.6172134
#> [2,] -0.5773503  0.1543033
#> [3,] -0.5773503 -0.7715167
#> 
#> $r
#>           [,1]      [,2]
#> [1,] -1.732051 -5.773503
#> [2,]  0.000000 -2.160247
# checks if result is correct
# should be x
qr$q %*% qr$r
#>      [,1] [,2]
#> [1,]    1    2
#> [2,]    1    3
#> [3,]    1    5
# should be identity matrix
t(qr$q) %*% qr$q
#>      [,1] [,2]
#> [1,]    1    0
#> [2,]    0    1

This is a example showing how to use QREcon when result in stored in matrices passed to the routine:

library(QREcon)
# same matrix as in other example
x <- matrix(c(1.,1.,1.,2.,3.,5.), 3, 2)
# allocate memory for q and r
q <- matrix(0., 3, 2)
r <- matrix(0., 2, 2)
# call QR decomposition routine
QREcon(x, q, r)
#> [1] 0
# display q and r. Should be same as above.
q
#>            [,1]       [,2]
#> [1,] -0.5773503  0.6172134
#> [2,] -0.5773503  0.1543033
#> [3,] -0.5773503 -0.7715167
r
#>           [,1]      [,2]
#> [1,] -1.732051 -5.773503
#> [2,]  0.000000 -2.160247

jimb3/QREcon documentation built on May 7, 2019, 6:08 a.m.