plot3darcs function

Modes of variation plots for PCA and PNSS

Modes of variation plots for PCA and PNSS based on 3D views and arcs along a mode. c * sd : the extent along lower and upper principal arcs.

The lower principal arc -> 0 -> upper principal arc has a total of 2*nn+1 configurations with: nn configurations along the negative principal arc to 0; one configuration at the PNS mean; nn configurations along the positive principal arc.

plot3darcs(x,pcno=1,c=1,nn=100,boundary.data=TRUE,view.theta=0,view.phi=0,type="pnss")

Arguments

• x: Output from pnss3d
• pcno: The number of the PC/PNSS component. The default is 1, the first PC/PNSS
• c: Number of standard deviations along each arc
• nn: In total 2 * nn + 1 configurations: n configurations on arc from negative to 0; 1 configuration at 0; nn configurations from 0 to positive
• boundary.data: Logical for whether to use boundary data or not.
• view.theta: Viewing angle theta
• view.phi: Viewing angle phi
• type: "pnss" principal nested sphere mean and arc, or "pca" Procrustes mean and linear PC.

Returns

A list with components - PNSmean: the PNSS mean

• lu.arc: the configurations along the arc

References

Dryden, I.L., Kim, K., Laughton, C.A. and Le, H. (2019). Principal nested shape space analysis of molecular dynamics data. Annals of Applied Statistics, 13, 2213-2234.

Jung, S., Dryden, I.L. and Marron, J.S. (2012). Analysis of principal nested spheres. Biometrika, 99, 551-568.

Author(s)

Kwang-Rae Kim, Ian Dryden

See Also

pns, pns4pc, pnss3d

Examples

ans <- pnss3d(digit3.dat, sphere.type="BIC", n.pc=5)
#aa <- plot3darcs(ans,c=2,pcno=1)
#bb <- plot3darcs(ans,c=2,pcno=1,type="pca")