inc {assist} | R Documentation |
Fit a Monotone Curve Using a Cubic Spline
Description
Return a spline fit of a increasing curve.
Usage
inc(y, x, spar = "v", limnla = c(-6, 0), grid = x, prec = 1e-06, maxit = 50, verbose = F)
Arguments
y |
a vecetor, used as the response data |
x |
a vector, used as the covariate. Assume an increasing relationshop of |
spar |
a character string specifying a method for choosing the smoothing parameter. "v", "m" and "u" represent GCV, GML and UBR respectively. Default is "v" for GCV |
limnla |
a vector of length one or two, specifying a search range for log10(n*lambda), where lambda is the smoothing parameter and n is the sample size. If it is a single value, the smoothing parameter will be fixed at this value. |
grid |
a vector of |
prec |
a numeric value used to assess convergence. Default is 1e-6 |
maxit |
an integer represeenting the maximum iterations. Default is 50. |
verbose |
an optional logical value. If ‘TRUE’, detailed iteration results are displayed. Default is "FALSE" |
Details
This function is to fit a increasing fucntion to the data. The monotone function is expressed as integral of an unknown function that a cubic spline is used to estimate.
Value
a split fit together with the convergence information
Author(s)
Yuedong Wang yuedong@pstat.ucsb.edu and Chunlei Ke chunlei_ke@yahoo.com
See Also
ssr