SVMR

SVM regression

       W = SVMR(X,NU,KTYPE,KPAR,EP)
       W = X*SVMR([],NU,KTYPE,KPAR,EP)
       W = X*SVMR(NU,KTYPE,KPAR,EP)

Input
X	Regression dataset
NU	Fraction of objects outside the 'data tube'
KTYPE	Kernel type (default KTYPE='p', for polynomial)
KPAR	Extra parameter for the kernel
EP	Epsilon, with of the 'data tube'

Output
W	Support vector regression

Description

Train an nu-Support Vector Regression on dataset X with parameter NU.  The kernel is defined by kernel type KTYPE and kernel parameter KPAR.  For the definitions of these kernels, have a look at proxm.m.

See also

linearr, proxm, gpr, svc,