.BG
.FN l1fit
.TL
Minimum Absolute Residual (L1) Regression
.CS
l1fit(x, y, int=TRUE)
.EQ
delim $$
.EN
.AG x
$X$ matrix for fitting $Y~=~Xb~+~e$ with variables in columns,
observations across rows.  Should not contain column of 1's
(see argument `int').  Number of rows of `x' should equal
the number of data values in `y'.  There should be fewer
columns than rows.
.AG y
numeric vector with as many observations as the number of
rows of `x'.
.AG int
flag for intercept; if `TRUE' an intercept term is
included in regression model.
.RT
list defining the regression (compare function
`lsfit').
.SP
.RC coef
vector of coefficients with constant term if `int' is `TRUE'.
.RC resid
residuals from the fit, i.e. `resid' is $Y~-~Xb$.
.SH REFERENCE
Barrodale and Roberts,
"Solution of an Overdetermined System of Equations in the L1 Norm",
.ul
CACM,
June 1974, pp. 319-320.
.KW array
.KW regression
.EQ
delim off
.EN
.WR
