tests/testthat/test.dredge_correlation.r

In markanewman/mndredge: Data dredging in R

context("dredge_correlation")

library(magrittr)
#library(testthat)

rm(list = ls())
set.seed(0)

seqcol <- seq(-10, 10, .5) %>% as.integer()
seqcol_tosmall <- 1:2
noisecol <- seqcol %>% length() %>% rnorm()
sncol <- seqcol + noisecol
s3ncol <- seqcol^3 + noisecol
charcol <- rep('a', seqcol %>% length())
empty_grid <- 
  data.frame(
    pair_1 = character(),
    pair_2 = character(),
    lag = integer())

#### parameters invalid on its face ####

# bad df
expect_error(
  NULL %>%
    dredge_correlation(
      grid = empty_grid,
      min_pvalue = 0,
      max_pvalue = .05),
  "data %>% is.data.frame()",
  fixed = T)

# not enough rows
expect_error(
  data.frame(seqcol_tosmall) %>%
    dredge_correlation(
      grid = empty_grid,
      min_pvalue = 0,
      max_pvalue = .05),
  "data %>% nrow() >= 3",
  fixed = T)

# bad grid
expect_error(
  data.frame(seqcol) %>%
    dredge_correlation(
      grid = NULL,
      min_pvalue = 0,
      max_pvalue = .05),
  "grid %>% is.data.frame()",
  fixed = T)

# bad min_pvalue
expect_error(
  data.frame(seqcol) %>%
    dredge_correlation(
      grid = empty_grid,
      min_pvalue = "a",
      max_pvalue = 0.05),
  "min_pvalue %>% is.numeric()",
  fixed = T)

# min_pvalue too small
expect_error(
  data.frame(seqcol) %>%
    dredge_correlation(
      grid = empty_grid,
      min_pvalue = -1,
      max_pvalue = 0.05),
  "min_pvalue >= 0",
  fixed = T)

# min_pvalue too big
expect_error(
  data.frame(seqcol) %>%
    dredge_correlation(
      grid = empty_grid,
      min_pvalue = 1,
      max_pvalue = 0.05),
  "min_pvalue < 1",
  fixed = T)

# bad max_pvalue
expect_error(
  data.frame(seqcol) %>%
    dredge_correlation(
      grid = empty_grid,
      min_pvalue = 0,
      max_pvalue = "a"),
  "max_pvalue %>% is.numeric()",
  fixed = T)

# max_pvalue too small
expect_error(
  data.frame(seqcol) %>%
    dredge_correlation(
      grid = empty_grid,
      min_pvalue = 0,
      max_pvalue = -1),
  "max_pvalue >= 0",
  fixed = T)

# max_pvalue too big
expect_error(
  data.frame(seqcol) %>%
    dredge_correlation(
      grid = empty_grid,
      min_pvalue = 0,
      max_pvalue = 1),
  "max_pvalue < 1",
  fixed = T)

# bad pvalue range
expect_error(
  data.frame(seqcol) %>%
    dredge_correlation(
      grid = empty_grid,
      min_pvalue = .5,
      max_pvalue = .25),
  "min_pvalue < max_pvalue",
  fixed = T)

#### normal analysis ####

# empty grid
data <- data.frame(a = seqcol, b = sncol)
actual <-
  data %>%
  dredge_correlation(
    grid = empty_grid,
    min_pvalue = 0,
    max_pvalue = .05)
expect_equal(actual %>% nrow(), 0)

# simple 2 column test
data <- data.frame(a = seqcol, b = sncol)
grid <- data %>% make_correlation_grid(max_lag = 0)
actual <-
  data %>%
  dredge_correlation(
    grid = grid,
    min_pvalue = 0,
    max_pvalue = .05)
expected <-
  data.frame(
    column_a = c("a"),
    column_b = c("b"),
    estimate = c(0.9882),
    p.value = c(0),
    lag = c(0),
  stringsAsFactors = F)
expect_equal(actual, expected, tolerance = 0.001)

# filter out poor correlation
data <- data.frame(a = seqcol, b = noisecol)
grid <- data %>% make_correlation_grid(max_lag = 0)
actual <-
  data %>%
  dredge_correlation(
    grid = grid,
    min_pvalue = 0,
    max_pvalue = .05)
expect_equal(actual %>% nrow(), 0)

# simple 2 column test + lag
data <- data.frame(a = seqcol, b = sncol)
grid <- data %>% make_correlation_grid(max_lag = 1)
actual <-
  data %>%
  dredge_correlation(
    grid = grid,
    min_pvalue = 0,
    max_pvalue = .05)
expected <-
  data.frame(
    column_a = c("a", "a"),
    column_b = c("b", "b"),
    estimate = c(0.9882, 0.9839),
    p.value = c(0, 0),
    lag = c(0, 1),
    stringsAsFactors = F)
expect_equal(actual, expected, tolerance = 0.001)

# simple 3 column test
data <- data.frame(a = seqcol, b = sncol, c = s3ncol)
grid <- data %>% make_correlation_grid(max_lag = 0)
actual <-
  data %>%
  dredge_correlation(
    grid = grid,
    min_pvalue = 0,
    max_pvalue = .05)
expected <-
  data.frame(
    column_a = c("a", "a", "b"),
    column_b = c("b", "c", "c"),
    estimate = c(0.9882, 0.9136, 0.9053),
    p.value = c(0, 0, 0),
    lag = c(0, 0, 0),
    stringsAsFactors = F)
expect_equal(actual, expected, tolerance = 0.001)

# 3 column test + one poor column
data <- data.frame(a = seqcol, b = noisecol, c = s3ncol)
grid <- data %>% make_correlation_grid(max_lag = 0)
actual <-
  data %>%
  dredge_correlation(
    grid = grid,
    min_pvalue = 0,
    max_pvalue = .05)
expected <-
  data.frame(
    column_a = c("a"),
    column_b = c("c"),
    estimate = c(0.9136),
    p.value = c(0),
    lag = c(0),
    stringsAsFactors = F)
expect_equal(actual, expected, tolerance = 0.001)