dev/tests/utils-test.R

In meantrix/corrp: Compute Correlations Type Analysis in Parallel

library("corrp")


df <- iris
n.cores <- 1
p.value <- 0.05
verbose <- TRUE
comp <- "g"
alternative <- "g"
cor.nn <- "pps"
cor.nc <- "pps"
cor.cc <- "pps"
parallel <- TRUE
n.sum <- 500
rk <- F
lm.args <- lm.args <- pearson.args <- cramersV.args <- dcor.args <- pps.args <- mic.args <- uncoef.args <- list()


# test pps unit
for (i in 1:NCOL(df)) {
  for (j in 1:NCOL(df)) {
    p <- corr_fun(df, ny = colnames(df)[i], nx = colnames(df)[j], cor.nn = cor.nn, cor.nc = cor.nc, cor.cc = cor.cc)
    p2 <- ppsr::score(df, y = colnames(df)[i], x = colnames(df)[j])
    # browser()  # check outputs
  }
}

p <- corrp(df, cor.nn = cor.nn, cor.nc = cor.nc, cor.cc = cor.cc)
p <- corr_matrix(p, cor.nn = cor.nn, cor.nc = cor.nc, cor.cc = cor.cc)
# ptest

x <- iris[[1]]
y <- iris[[2]]

x1 <- ptest(x, y, FUN = function(x, y) cor(x, y), num.s = 2000, alternative = "t")
x2 <- ptest(x, y, FUN = function(x, y) cor(x, y), num.s = 2000, alternative = "g")
x3 <- ptest(x, y, FUN = function(x, y) cor(x, y), num.s = 2000, alternative = "l")

y1 <- cor.test(x, y, alternative = "t")$p.value
y2 <- cor.test(x, y, alternative = "g")$p.value
y3 <- cor.test(x, y, alternative = "l")$p.value

x4 <- sample(1:30, 100, replace = TRUE)
y4 <- sample(0:1, 100, replace = TRUE)