Monte Carlo rank test

Computes a rank-based Monte Carlo p-value with the standard +1 correction. The helper is intended for comparing a scalar observed statistic with the corresponding statistics from simulated trajectories.

Usage

mc_rank_test(
  observed,
  simulated,
  alternative = c("two.sided", "less", "greater")
)

Arguments

observed

Numeric scalar observed statistic.

simulated

Numeric vector of simulated statistics.

alternative

Character. One of "two.sided", "less", or "greater". For "greater", small p-values indicate that observed is unusually large relative to simulated. For "less", small p-values indicate that observed is unusually small.

Value

A list with observed, simulated, rank, p_value, and alternative.

References

North, B. V., Curtis, D., and Sham, P. C. (2002). A note on the calculation of empirical P values from Monte Carlo procedures. American Journal of Human Genetics, 71(2), 439-441. https://doi.org/10.1086/341527

Examples

mc_rank_test(5, c(1, 2, 3, 4), alternative = "greater")
#> $observed
#> [1] 5
#> 
#> $simulated
#> [1] 1 2 3 4
#> 
#> $rank
#> [1] 5
#> 
#> $p_value
#> [1] 0.2
#> 
#> $alternative
#> [1] "greater"
#> 
#> attr(,"class")
#> [1] "gmov_mc_rank_test" "list"