R/random.equal.R

In rportfolios: Random Portfolio Generation

random.equal <- function( n=2, k=n, segments = NULL, x.t=1 )
{
###
### This function returns an n by 1 vector of investment weights where
### there are only k non-zero weights each of which is x.t / k
###
### Arguments
### n = a positive integer value for the number of investments in the portfolio
### k = a positive integer value for the number of non-zero investments
### x.t = a positive numeric value for the sum of the weights
###
    if ( k > n )
       stop( "argument 'k' is greater than 'n'" )
    if ( is.null( segments ) ) {
        weight <- x.t / k
        investments <- sample( 1:n, k, replace=FALSE )
    }
    else {
        segmentInvestments <- collapse.segments( segments )
        numberInvestments <- length( segmentInvestments )
        if ( numberInvestments > n || max( segmentInvestments ) > n )
            stop( "argument 'segments' has investments that are not allowed" )
        weight <- x.t / numberInvestments
        investments <- sample( segmentInvestments, numberInvestments, replace = FALSE )
    }    
    x <- rep( 0, n )
    x[investments] <- weight
    return( x )
}