GPArotation (version 2014.11-1)

Random.Start: Generate a Random Orthogonal Rotation

Description

Random orthogonal rotation to use as Tmat matrix to start GPForth or GPFoblq.

Usage

Random.Start(k)

Arguments

k

An integer indicating the dimension of the square matrix.

Value

An orthogonal matrix.

Details

The random start function produces an orthogonal matrix with columns of length one based on the QR decompostion.

See Also

GPForth, GPFoblq, oblimin

Examples

# NOT RUN {
   Global.min <- function(A,method,B=10){
      fv <- rep(0,B)
      seeds <- sample(1e+7, B)
      for(i in 1:B){
    	cat(i," ")
    	set.seed(seeds[i])
    	gpout <- GPFoblq(A=A, Random.Start(ncol(A)), method=method)
    	dtab <- dim(gpout$Table)
    	fv[i] <- gpout$Table[dtab[1],2]
    	cat(fv[i], "\n")
      }
      cat("Min is ",min(fv),"\n")
      set.seed(seeds[order(fv)[1]])
      ans <- GPFoblq(A=A, Random.Start(ncol(A)), method=method)
      ans
      }

   data("Thurstone", package="GPArotation")

   Global.min(box26,"simplimax",10)
   
# }

Run the code above in your browser using DataCamp Workspace