tests/testthat/test_data_glm.r

In simglm: Simulate Models Based on the Generalized Linear Model

# context('data_glm')
# 
# test_that('probabilities between 0 and 1', {
#   Xmat <- cbind(rnorm(50), rnorm(50), rnorm(50))
#   beta <- c(4, 3, 1)
#   expect_gte(min(data_glm_single(Xmat, beta, 50, 
#                                  outcome_type = 'logistic')$logistic), 0)
#   expect_lte(max(data_glm_single(Xmat, beta, 50, 
#                                  outcome_type = 'logistic')$logistic), 1)
#   
#   # z_mat <- Xmat[, 1]
#   # rand_eff <- c(rnorm(50))
#   # expect_gte(min(data_glm_nested(Xmat, z_mat, beta, rand_eff, 10, 5)$logistic), 0)
# })
# 
# test_that('correct length', {
#   Xmat <- cbind(rnorm(50), rnorm(50), rnorm(50))
#   beta <- c(4, 3, 1)
#   expect_equal(nrow(data_glm_single(Xmat, beta, 50, outcome_type = 'logistic')), 50)
# })
# 
# test_that('sim_data are 0 or 1s inclusive', {
#   Xmat <- cbind(rnorm(50), rnorm(50), rnorm(50))
#   beta <- c(4, 3, 1)
#   expect_length(table(data_glm_single(Xmat, beta, 50, outcome_type = 'logistic')$sim_data), 2)
# })
# 
# context('data_reg')
# 
# test_that('correct length', {
#   Xmat <- cbind(rnorm(50), rnorm(50), rnorm(50))
#   beta <- c(4, 3, 1)
#   err <- rnorm(50)
#   expect_equal(nrow(data_reg_single(Xmat, beta, 50, err)), 50)
# })
# 
# test_that('Fbeta + err = sim_data', {
#   Xmat <- cbind(rnorm(50), rnorm(50), rnorm(50))
#   beta <- c(4, 3, 1)
#   err <- rnorm(50)
#   tmp <- data_reg_single(Xmat, beta, 50, err)
#   expect_equal(tmp[, 1] + tmp[, 2], tmp[, 3])
# })
# 
#