# wren tests context("wrens") test_that("wren 5 minute counts works",{ ## standard five-minute counts data(wren_5min) cu_wren_5min <- 1/sqrt(10000) w1_df_unif <- ds(wren_5min, transect="point", truncation=110, key="unif", convert_units=cu_wren_5min, order=c(1,2), er_var="P3", optimizer = "R") # do the same thing with dht2 w1_nhat <- dht2(w1_df_unif, flatfile=wren_5min, strat_formula=~Region.Label, convert_units=cu_wren_5min, er_est="P3") d_stuff <- w1_df_unif$dht$individuals$D expect_equal(attr(w1_nhat,"density")$Density, d_stuff$Estimate, tol=1e-3) expect_equal(attr(w1_nhat,"density")$LCI, d_stuff$lcl, tol=1e-3) expect_equal(attr(w1_nhat,"density")$UCI, d_stuff$ucl, tol=1e-3) expect_equal(attr(w1_nhat,"density")$Density_se, d_stuff$se, tol=1e-3) expect_equal(attr(w1_nhat,"density")$Density_CV, d_stuff$cv, tol=1e-3) }) test_that("wren snapshot works",{ ## the ‘snapshot’ method data(wren_snapshot) cu_wren_snapshot <- 1/sqrt(10000) w2_df_hr <- ds(wren_snapshot, transect="point", truncation=110, key="hr", adjustment=NULL, convert_units=cu_wren_snapshot, er_var="P3") w2_nhat <- dht2(w2_df_hr, flatfile=wren_snapshot, strat_formula=~Region.Label, convert_units=cu_wren_snapshot, er_est="P3") d_stuff <- w2_df_hr$dht$individuals$D expect_equal(attr(w2_nhat,"density")$n, w2_df_hr$dht$individuals$summary$n) expect_equal(attr(w2_nhat,"density")$k, w2_df_hr$dht$individuals$summary$k[1]) expect_equal(attr(w2_nhat,"density")$Effort, w2_df_hr$dht$individuals$summary$Effort) expect_equal(attr(w2_nhat,"density")$Covered_area, w2_df_hr$dht$individuals$summary$CoveredArea) expect_equal(attr(w2_nhat,"density")$Density, d_stuff$Estimate, tol=1e-3) expect_equal(attr(w2_nhat,"density")$LCI, d_stuff$lcl, tol=1e-3) expect_equal(attr(w2_nhat,"density")$UCI, d_stuff$ucl, tol=1e-3) expect_equal(attr(w2_nhat,"density")$Density_se, d_stuff$se, tol=1e-6) expect_equal(attr(w2_nhat,"density")$Density_CV, d_stuff$cv, tol=1e-3) }) test_that("wren cue count works",{ ## cue count method data(wren_cuecount) mult <- unique(wren_cuecount[, c("Cue.rate","Cue.rate.SE")]) names(mult) <- c("rate", "SE") # search time is the effort wren_cuecount$Effort <- wren_cuecount$Search.time # for comparability with distance, we ignore replicate visits wren_cuecount$Sample.Label_old <- wren_cuecount$Sample.Label wren_cuecount$Sample.Label <- sub("-\\d+", "", wren_cuecount$Sample.Label) cu <- sqrt(0.0001) w3_df_hr <- ds(wren_cuecount, transect="point", truncation=92.5, adjustment=NULL, key="hr", er_var="P3", convert_units=cu) w3_nhat <- dht2(w3_df_hr, flatfile=wren_cuecount, strat_formula=~Region.Label, multipliers=list(creation=mult), convert_units=cu, er_est="P3") expect_equal(w3_nhat$Abundance_se, 8.0002, tol=1e-2) expect_equal(attr(w3_nhat,"density")$Density, 1.2123, tol=1e-1) expect_equal(attr(w3_nhat,"density")$Density_se, 0.24262, tol=1e-2) expect_equal(attr(w3_nhat,"density")$Density_CV, .2001, tol=1e-2) expect_equal(attr(w3_nhat,"density")$LCI, 0.82134, tol=1e-2) expect_equal(attr(w3_nhat,"density")$UCI, 1.7893, tol=1e-2) }) #test_that("wren 4 works",{ # ## line transect data # data(wren4) # # cu <- (1/1000)*1/0.01 # # cribbed the best model # w4_df_hn <- ds(wren4, transect="line", truncation=95, key="hn", # adjustment=NULL, convert.units=cu) # # # result <- ddf(dsmodel = ~mcds(key = "hn", formula = ~1, adj.series="herm", adj.order=c(4,6)), # data = w4_df_hn$ddf$data, method = "ds", # meta.data = list(width = 95), # control=list(initial=list(scale=log(.3532E5), adjust=c(.1184E6, .1534E5)), # nofit=TRUE)) # # # w4_nhat <- dht2(result, flatfile=wren4, strat_formula=~Region.Label, # convert_units=cu) # expect_equal(attr(w4_nhat,"density")$Density, 2.1344, tol=1e-1) # expect_equal(attr(w4_nhat,"density")$Density_se, 0.21949, tol=1e-2) # expect_equal(attr(w4_nhat,"density")$Density_CV, .1028, tol=1e-2) # expect_equal(attr(w4_nhat,"density")$LCI, 1.7382, tol=1e-2) # expect_equal(attr(w4_nhat,"density")$UCI, 2.6207, tol=1e-2) #})