context("Chapter 6") test_that("Chapter 6 functions basically work", { n_short <- floor(lydersen_2012a / 2) expect_output( object = print(Brant_test_2xc(lydersen_2012a)), regexp = "P = 0.434217, T = 1.668 \\(df = 2\\)" ) expect_output( object = print(Cumulative_models_for_2xc(lydersen_2012a)), regexp = "Wald \\(Z-statistic\\): P = 0.14097, Z = -1.472" ) expect_output( object = print(Cumulative_models_for_2xc(fontanella_2008, "probit")), regexp = " Pearson .+ P = 0.06041, X2 = 5.613 \\(df=2\\)" ) expect_output( object = print(Cumulative_models_for_2xc(fontanella_2008, "logit")), regexp = "Wald \\(OR\\) 0.318 0.228 to 0.444" ) expect_error(Cumulative_models_for_2xc(tea), "must have at least 3 columms") expect_error(Cumulative_models_for_rxc(n_short[, 1, drop = FALSE]), "must have at least 3") expect_error(Cumulative_models_for_rxc(n_short[, 1:2]), "must have at least 3") expect_output( object = print(Exact_cond_midP_linear_rank_tests_2xc(n_short)), regexp = "Exact cond. linear rank test: P = 0.23854" ) expect_output( object = print( Exact_cond_midP_linear_rank_tests_2xc(matrix(c(1, 200, 300, 0, 5, 6), 2)) ), "Exact cond. linear rank test: P = 0.00000" ) dir <- "decreasing" expect_output( object = print(Exact_cond_midP_unspecific_ordering_rx2(t(n_short), dir)), regexp = "Exact conditional test\\s*:\\s*P =\\s*0.23094" ) stat <- "PearsonCumOR" expect_output( object = print(Exact_cond_midP_unspecific_ordering_rx2(t(n_short), dir, stat)), regexp = "Exact conditional test\\s*:\\s*P =\\s*0.08012" ) expect_error( Exact_cond_midP_unspecific_ordering_rx2(matrix(1:4, 2), "increasing"), "n must have either 4 or 5 rows. Consider transposing it." ) expect_output( object = print(MantelHaenszel_test_2xc(lydersen_2012a)), regexp = "test of association: P = 0.1442, T = 2.132 \\(df=1\\)" ) expect_output( object = print(Pearson_LR_tests_cum_OR_2xc(lydersen_2012a)), regexp = "Pearson chi-squared test: T = 3.813, P = 0.07223" ) expect_output( object = print(Pearson_LR_tests_cum_OR_2xc(t(lydersen_2012a), "increasing")), regexp = "Pearson chi-squared test: T = 2.708, P = 0.04993" ) alphahat0 <- c(-0.1923633, 0.5588396, 1.271953) expect_output( object = print( Score_test_for_effect_in_the_probit_model_2xc( lydersen_2012a, alphahat0 ) ), regexp = "Score test for effect: P = 0.1431, T = 2.145 \\(df=1\\)" ) expect_output( object = print(the_2xc_table(n_short, direction = "decreasing")), regexp = "Wald \\(OR\\) 2.420 0.598 to 9.788" ) })