This is the basic computing engine called by monpol used to fit
monotonic polynomials. These should usually not be used
directly unless by experienced users.
Usage
monpol.fit(x, y, w, K=1, start, trace = FALSE, plot.it = FALSE,
control = monpol.control(),
algorithm = c("Full", "Hawkins", "BCD", "CD1", "CD2"),
ptype = c("Elphinstone", "EHH", "Penttila"),
ctype = c("cge0", "c2"))
SOSpol.fit(x, y, w = NULL, deg.is.odd, K, start, a, b,
monotone = c("increasing", "decreasing"),
trace = FALSE, plot.it = FALSE, type,
control = monpol.control())
Arguments
x
vector containing the observed values for the regressor variable.
y
vector containing the observed values for the response
variable; should be of same length as x.
w
optional vector of weights; should be of the same length as
x if specified.
deg.is.odd, K
“deg.is.odd” is a logical, “K” is a
non negative integer. If “deg.is.odd” is TRUE then a
polynomial with highest power 2K+1 will be fitted to
the data, otherwise the highest order will be 2K.
start
optional starting value for the iterative fitting.
a,b, type
polynomial should be monotone on the interval from a
to b; “type” should be 0 if neither of the boundaries
is finite, 1 if a if finite but not b and 2 if both
boundaries are finite.
monotone
force the desired monotonicity in case the default
choice is wrong.
trace
print out information about the progress of the
interative fitting at the start and then every trace
iterations.
plot.it
plot the data and initial fit, then plot current fit
every plot.it iterations.
control
settings that control the iterative fit; see
monpol.control for details.
algorithm
algorithm to be used; see monpol for
details.
ptype
parameterisation to be used; see monpol for
details.
ctype
parameterisation to be used; see monpol for
details.
Value
a list with components
par
the fitted parameters.
grad
the gradient of the objective function at the fitted
parameters.
beta
the coefficients of the fitted polynomial in the
‘beta’ parameterisation; on the fitted scale.
RSS
the value of the objective function; on the fitted scale.
niter
number of iterations.
converged
indicates whether algorithm has converged.
ptype
input parameter ptype.
ctype
input parameter cptype.
beta.raw
the coefficients of the fitted polynomial in the
‘beta’ parameterisation; on the original scale.
fitted.values
the fitted values; on the fitted scale.
residuals
the residuals; on the fitted scale.
K
input parameter K.
minx
the minimum value in the vector x.
sclx
the difference between the maximum and minimum values in
the vector x.
miny
the minimum value in the vector y.
scly
the difference between the maximum and minimum values in
the vector y.
algorithm
input paramater algorithm.
Author(s)
Berwin A Turlach
References
Murray, K., M<c3><83><c2><bc>ller, S. and Turlach, B.A. (2016). Fast and
flexible methods for monotone polynomial fitting, Journal of
Statistical Computation and Simulation.
Accepted for publication, doi:10.1080/00949655.2016.1139582.
Murray, K., M<c3><83><c2><bc>ller, S. and Turlach,
B.A. (2013). Revisiting fitting monotone polynomials to data,
Computational Statistics28(5):
1989–2005. Doi:10.1007/s00180-012-0390-5.
See Also
monpol which you should use for fitting monotonic
polynomials unless you know better.