rrace.Rd
Returns a rank vector suitable for interpretation with race()
.
rrace(strengths)
Uses a simple recursive system to generate the ranks.
Returns a character vector with entries corresponding to the competitor. The first element is the winner, the second the runner-up, and so on, until the final element is the last to cross the finishing line.
o <- c(a=0.4, b=0.3, c=0.2, d=0.1)
rrace(o)
#> [1] "b" "a" "c" "d"
rankvec_likelihood(rrace(o))
#> log( a * (a + b + c + d)^-1 * (a + b + d)^-1 * (a + d)^-1 * b * c)
D <- t(replicate(10,rrace(o))) # 10 races
H <- hyper2()
for(i in seq_len(nrow(D))){H <- H+rankvec_likelihood(D[i,])}
maxp(H) # should be close to o
#> a b c d
#> 0.4717220 0.1638641 0.2510664 0.1133476