Compute (truncated) orthonormal eigenfunctions and scores for data potentially on a non-equidistant grid.

fpc_wsvd(data, arg, pve = 0.995)

Arguments

data

numeric matrix of function evaluations (each row is one curve, no NAs)

arg

numeric vector of argument values

pve

percentage of variance explained

Value

a list with entries

  • `mu`` estimated mean function (numeric vector)

  • `efunctions`` estimated FPCs (numeric matrix, columns represent FPCs)

  • scores estimated FPC scores (one row per observed curve)

  • npc how many FPCs were returned for the given pve (integer)

References

code adapted from / inspired by wsvd() function of Bioconductor package mogsa by Cheng Meng.