context("check accuracy of rflex.test poisson") set.seed(15) data(nydf) data(nyw) outp <- rflex.test( coords = cbind(nydf$longitude, nydf$latitude), cases = floor(nydf$cases), pop = nydf$population, w = nyw, k = 50, nsim = 99, alpha = 1, longlat = FALSE, alpha1 = 0.2 ) # results taken from rflex_test_ny_poisson_50nn_cartesian # compare with FlexScan original for NY data test_that("check accuracy for rflex.test poisson", { expect_equal( sort(outp$clusters[[1]]$locids), c(1:2, 13, 15, 27, 35, 37:38, 43, 46:47, 49, 51:53) ) expect_equal(round(outp$clusters[[1]]$max_dist, 5), 0.22483) expect_equal(outp$clusters[[1]]$cases, 79) expect_equal(round(outp$clusters[[1]]$exp, 4), 36.1025) expect_equal(round(outp$clusters[[1]]$smr, 5), 2.18822) expect_equal(round(outp$clusters[[1]]$loglik, 4), 20.8014) # p-values are tough to test, make sure results don't change # in future versions since these were manually checked expect_equal(outp$clusters[[1]]$pvalue, 0.01) expect_equal(sort(outp$clusters[[2]]$locids), c(89, 90)) expect_equal(round(outp$clusters[[2]]$max_dist, 6), 0.125172) expect_equal(outp$clusters[[2]]$cases, 13) expect_equal(round(outp$clusters[[2]]$exp, 5), 3.46124) expect_equal(round(outp$clusters[[2]]$smr, 5), 3.75588) expect_equal(round(outp$clusters[[2]]$loglik, 5), 7.74784) # p-values are tough to test, make sure results don't change # in future versions since these were manually checked expect_equal(round(outp$clusters[[2]]$pvalue, 1), 0.3) expect_equal( sort(outp$clusters[[12]]$locids), c(135, 146, 208, 210) ) expect_equal(round(outp$clusters[[12]]$max_dist, 6), 0.030934) expect_equal(outp$clusters[[12]]$cases, 12) expect_equal(round(outp$clusters[[12]]$exp, 5), 5.59582) expect_equal(round(outp$clusters[[12]]$smr, 5), 2.14446) expect_equal(round(outp$clusters[[12]]$loglik, 5), 2.78814) # p-values are tough to test, make sure results don't change # in future versions since these were manually checked expect_equal(outp$clusters[[12]]$pvalue, 1) })