fda.usc (version 1.2.3)

gridfdata, rcombfdata: Utils for generate functional data

Description

gridfdata generates n curves as lineal combination of the original curves fdataobj plus a functional trend mu.

rcombfdata generates n random combinations of the fdataobj curves plus a functional trend mu. The coefficients of the combinations follows a normal distribution with zero mean and standard deviation sdarg.

Usage

gridfdata(coef,fdataobj,mu)
rcombfdata(n = 10, fdataobj, mu,  sdarg = rep(1,nrow(fdataobj)), norm = 1)

Arguments

coef

Coefficients of the combination. A matrix with number of columns equal to number of curves in fdataobj

fdataobj

fdata class object.

mu

Functional trend, by default mu=$\mu(t)=0$. An object of class fdata.

n

Number of curves to be generated

sdarg

Standard deviation of the coefficients.

norm

Norm of the coefficients. The norm is adjusted before the transformation for sdarg is performed.

Value

Return the functional trajectories as a fdata class object.

See Also

See Also as rproc2fdata

Examples

tt=seq(0,1,len=51)
fou3=create.fourier.basis(c(0,1),nbasis=3)
fdataobj=fdata(t(eval.basis(tt,fou3)),argvals=tt)

coef=expand.grid(0,seq(-1,1,len=11),seq(-1,1,len=11))
grid=gridfdata(coef,fdataobj)
plot(grid,lty=1)

rcomb=rcombfdata(n=51,fdataobj,mu=fdata(30*tt*(1-tt),tt))
plot(rcomb,lty=1)