inst/tinytest/test_dtsurvey.R
In PHSKC-APDE/dtsurvey: Survey Analysis Using data.table For SPEEEEEED
library('tinytest')
library('survey')
library('data.table')
library('dtsurvey')
#dumb survey
set.seed(98112)
size = 101
fake = data.table(num = runif(size),
fact = as.factor(sample(letters[1:5], size, TRUE)),
logi = sample(c(TRUE, FALSE), size, TRUE),
weight = 1,
strata = sample(1:10, size, TRUE),
byvar = sample(1:2, size, TRUE),
byna = sample(c(NA, 1, 2), size , TRUE))
fake[, psu := as.numeric(paste0(strata, sample(1:5, size, TRUE)))]
fake[, num_na := num]
fake[sample(seq_len(size), 20), num_na := NA]
fake[, fact_na := fact]
fake[sample(seq_len(size), 20), fact_na := NA]
levs = unique(fake[, fact])
fake[, paste0('fct_', levs) := lapply(levs, function(x) as.numeric(fact == x))]
fake[, paste0('fct_na_', levs) := lapply(levs, function(x) as.numeric(fact_na == x))]
fake[, bin := sample(0:1, size, T)]
fake[bin == 0, bin := NA]
fake[, bin_by := bin]
fake[, bin_fact := as.factor(bin)]
fake[!is.na(bin), bin2 := sample(0:1, .N, T)]
setorder(fake, byvar)
#make sure the survey package will take it
expect_silent(og <- survey::svydesign(~psu, strata = ~strata, data = fake, weights = ~weight), info = 'Fake data works with survey package')
fake_sur = dtsurvey(fake, psu = 'psu', strata = 'strata', weight = 'weight', nest = T)
#confirm that a mean, no bys or NAs are equal
r1.1 = svymean(~num, og)
r1.2 = fake_sur[, smean(num, var_type = 'se', ids = `_id`)]
expect_equal(as.numeric(r1.1), r1.2$result, info = 'Simple survey means are equal')
expect_equal(as.numeric(SE(r1.1)), r1.2$se, info = 'Simple survey ses are equal')
#this time with bys
r2.1 = svyby(~num, ~byvar, og, svymean)
r2.2 = fake_sur[, smean(num, var_type = 'se', ids = `_id`), by = byvar]
setorder(r2.2, byvar)
expect_equal(r2.1$num, r2.2$result, info = 'Simple survey means are equal, with one byvar')
expect_equal(as.numeric(SE(r2.1)), r2.2$se, info = 'Simple survey ses are equal, with one byvar')
#simple, with NA
r3.1 = svymean(~num_na, og, na.rm = T)
r3.2 = fake_sur[, smean(num_na, var_type = 'se', na.rm = TRUE, ids = `_id`)]
expect_equal(as.numeric(r3.1), r3.2$result, info = 'Simple survey means are equal, with some NAs')
expect_equal(as.numeric(SE(r3.1)), r3.2$se, info = 'Simple survey ses are equal, with some NAs')
#simple, with NA and by
r4.1 = svyby(~num_na, ~byvar, og, svymean, na.rm = TRUE)
r4.2 = fake_sur[, smean(num_na, var_type = 'se', ids = `_id`), by = byvar]
setorder(r4.2, byvar)
expect_equal(r4.1$num_na, r4.2$result, info = 'Simple survey means are equal, with one byvar and some NAs')
expect_equal(as.numeric(SE(r4.1)), r4.2$se, info = 'Simple survey ses are equal, with one byvar and some NAs')
#try assignments
expect_silent(r5 <- fake_sur[, blah := smean(num, ids = `_id`, var_type = NULL)], info = 'Simple survey mean assignment works')
expect_equal(r5[1, blah], as.numeric(svymean(~num, og, na.rm = T)))
expect_silent(r6 <- fake_sur[, blah := smean(num_na, ids = `_id`, na.rm = T, var_type = NULL), by = byvar], info = ' Simple survey mean assignment with by')
expect_equal(sort(unique(r6[, blah])), svyby(~num_na, ~byvar, og, svymean, na.rm = T)$num_na)
#means, se, degrees of freedom
expect_silent(r7.1 <- fake_sur[, smean(num, ids = `_id`, var_type = c('se', 'ci'), use_df = FALSE)])
r7.2 <- svymean(~num, og)
expect_equal(unname(unlist(r7.1)), unname(c(coef(r7.2), SE(r7.2), confint(r7.2))))
#Test alternative ways of calculating cis, first without by vars
r13.1 <- fake_sur[, smean(logi, ids = `_id`, var_type = 'ci', ci_method = 'xlogit')]
r13.2 <- fake_sur[, smean(logi, ids = `_id`, var_type = 'ci', ci_method = 'beta')]
r13.3 <- confint(svyciprop(~logi, og, method = 'xlogit', na.rm = T))
r13.4 <- confint(svyciprop(~logi, og, method = 'beta', na.rm = T))
expect_equal(unname(unlist(r13.1)[2:3]), as.vector(unname(r13.3)))
#alternative ci methods, with bys
r14.1 <- fake_sur[, smean(logi, ids = `_id`, var_type = 'ci', ci_method = 'xlogit'), byvar]
r14.2 <- fake_sur[, smean(logi, ids = `_id`, var_type = 'ci', ci_method = 'beta'), byvar]
r14.3 <- svyby(~logi, ~byvar, og, svyciprop, vartype = 'ci', method = 'xlogit', na.rm = T)
r14.4 <- svyby(~logi, ~byvar, og, svyciprop, vartype = 'ci', method = 'beta', na.rm = T)
expect_equal(unname(as.matrix(r14.1)), unname(as.matrix(r14.3)))
expect_equal(unname(as.matrix(r14.2)), unname(as.matrix(r14.4)))
#alternative ci methods, with filtering
r15.1 <- fake_sur[fact != 'a', smean(logi, ids = `_id`, var_type = 'ci', ci_method = 'xlogit'), byvar]
r15.2 <- fake_sur[fact != 'a', smean(logi, ids = `_id`, var_type = 'ci', ci_method = 'beta'), byvar]
r15.3 <- svyby(~logi, ~byvar, subset(og, fact != 'a'), svyciprop, vartype = 'ci', method = 'xlogit', na.rm = T)
r15.4 <- svyby(~logi, ~byvar, subset(og,fact != 'a'), svyciprop, vartype = 'ci', method = 'beta', na.rm = T)
expect_equal(unname(as.matrix(r15.1)), unname(as.matrix(r15.3)))
expect_equal(unname(as.matrix(r15.2)), unname(as.matrix(r15.4)))
#factors
r16.1 <- fake_sur[, smean(fact, ids = `_id`, var_type = 'se')]
r16.2 <- svymean(~fact, og, vartype = c('se'))
expect_equal(r16.1[, result], unname(coef(r16.2)))
expect_equal(r16.1[, se], unname(SE(r16.2)))
#factors, assign-- factor assignment is too annoying to get it to work. It probably needs to be wrapped in something else
# expect_silent(fake_sur[, blah_fact := smean(fact, ids = `_id`)])
# expect_true(all(unique(fake_sur[, blah_fact])[[1]] == fake_sur[, smean(fact, ids = `_id`)]))
#factors, by
r17.1 <- fake_sur[, smean(fact, ids = `_id`, var_type = c('se', 'ci'), use_df = FALSE), byvar]
r17.2 <- as.data.table(svyby(~fact, ~byvar, og, svymean))
r17.2 <- melt(r17.2, id.vars = 'byvar')
setorder(r17.2, byvar, variable)
expect_equal(r17.1[,result], r17.2[!substr(variable, 1,2) == 'se', value])
expect_equal(r17.1[,se], r17.2[substr(variable, 1,2) == 'se', value])
r17.3 = confint(svyby(~fact, ~byvar, og, svymean))
r17.1[, fact := rep(levels(fake_sur[, fact]),2)]
setorder(r17.1, fact, byvar)
expect_equal(unique(r17.1[,lower]), unname(r17.3[,1]))
expect_equal(unique(r17.1[,upper]), unname(r17.3[,2]))
#byna
r18.1 = fake_sur[, smean(num_na, ids = `_id`, var_type = 'se', use_df = FALSE), byna]
r18.2 = svyby(~num_na, ~byna, og, svymean, na.rm = T)
expect_false(length(r18.1[, result]) == length(r18.2[,2]), info = 'dtsurvey returns a row when a by var is NA, survey does not')
expect_equal((r18.1[!is.na(byna),result, keyby = byna][, result]), r18.2[,2], info = 'dtsurvey returns a row when a by var is NA, survey does not')
#multiby
r19.1 = fake_sur[, .(result = smean(num, ids = `_id`)), keyby = .(byvar, logi)] #todo, why does this return V1?
r19.2 = svyby(~num, ~byvar + logi, og, svymean, na.rm = T)
setorder(r19.1, logi, byvar)
expect_equal(r19.1[, result], r19.2[, 'num'])
#confirm that a total, no bys or NAs are equal
r20.1 = svytotal(~num, og)
r20.2 = fake_sur[, stotal(num, var_type = 'se', ids = `_id`)]
expect_equal(as.numeric(r20.1), r20.2$result, info = 'Simple rep survey totals are equal')
expect_equal(as.numeric(SE(r20.1)), r20.2$se, info = 'Simple rep totals ses are equal')
#this time with bys
r21.1 = svyby(~num, ~byvar, og, svytotal)
r21.2 = fake_sur[, stotal(num, var_type = 'se', ids = `_id`), by = byvar]
setorder(r21.2, byvar)
expect_equal(r21.1$num, r21.2$result, info = 'Simple rep survey totals are equal, with one byvar')
expect_equal(as.numeric(SE(r21.1)), r21.2$se, info = 'Simple rep survey ses are equal, with one byvar')
#simple, with NA
r22.1 = svytotal(~num_na, og, na.rm = T)
r22.2 = fake_sur[, stotal(num_na, var_type = 'se', na.rm = TRUE, ids = `_id`)]
expect_equal(as.numeric(r22.1), r22.2$result, info = 'Simple rep survey totals are equal, with some NAs')
expect_equal(as.numeric(SE(r22.1)), r22.2$se, info = 'Simple rep survey ses are equal, with some NAs')
#simple, with NA and by
r23.1 = svyby(~num_na, ~byvar, og, svytotal, na.rm = TRUE)
r23.2 = fake_sur[, stotal(num_na, var_type = 'se', ids = `_id`), by = byvar]
setorder(r23.2, byvar)
expect_equal(r23.1$num_na, r23.2$result, info = 'Simple rep survey totals are equal, with one byvar and some NAs')
expect_equal(as.numeric(SE(r23.1)), r23.2$se, info = 'Simple rep survey ses are equal, with one byvar and some NAs')
#factors, by
r24.1 <- fake_sur[, stotal(fact, ids = `_id`, var_type = c('se', 'ci'), use_df = FALSE), keyby = byvar]
r24.2 <- as.data.table(svyby(~fact, ~byvar, og, svytotal))
r24.2 <- melt(r24.2, id.vars = 'byvar')
setorder(r24.2, byvar, variable)
expect_equal(r24.1[, result], r24.2[!substr(variable, 1,2) == 'se', value])
expect_equal(r24.1[,se], r24.2[substr(variable, 1,2) == 'se', value])
r24.3 = confint(svyby(~fact, ~byvar, og, svytotal))
r24.1[, fact := rep(levels(fake_sur[, fact]),2)]
setorder(r24.1, fact, byvar)
expect_equal(unique(r24.1[,lower]), unname(r24.3[,1]))
expect_equal(unique(r24.1[,upper]), unname(r24.3[,2]))
#converting a survey object into a dtsurvey (see more below)
r25 = as.dtsurvey(og)
expect_equivalent(r25, dtsurvey(fake, psu = 'psu', strata = 'strata', weight = 'weight', nest = T))
r26 = survey::svydesign(~1, strata = ~strata, data = fake, weights = ~weight, nest = T)
expect_equivalent(as.dtsurvey(r26), dtsurvey(fake, psu = NULL, strata = 'strata', weight = 'weight', nest = T))
#ci method errors
#Proportion methods for factors
#x logit
r27.1 = fake_sur[, smean(fact, var_type = c('ci'), ci_method = 'xlogit')]
r27.1 = data.table(r27.1)
r27.2 = lapply(levs, function(x){
r = svyciprop(as.formula(paste0('~','fct_',x, collapse = '')),og, 'xlogit')
r= c(r, confint(r))
r = data.table(t(r))
setnames(r, c('result', 'lower', 'upper'))
})
r27.2 = rbindlist(r27.2)[, levels := as.character(levs)]
setorder(r27.2, levels)
expect_equivalent(r27.1, r27.2)
#beta
r28.1 = fake_sur[, smean(fact, var_type = c('ci'), ci_method = 'beta')]
r28.1 = data.table(r28.1)
r28.2 = lapply(levs, function(x){
r = svyciprop(as.formula(paste0('~','fct_',x, collapse = '')),og, 'beta')
r= c(r, confint(r))
r = data.table(t(r))
setnames(r, c('result', 'lower', 'upper'))
})
r28.2 = rbindlist(r28.2)[, levels := as.character(levs)]
setorder(r28.2, levels)
expect_equivalent(r28.1, r28.2)
#with NAs
r29.1 = fake_sur[, smean(fact_na, var_type = c('ci'), ci_method = 'xlogit')]
r29.1 = data.table(r29.1)
r29.2 = lapply(levs, function(x){
r = svyciprop(as.formula(paste0('~','fct_na_',x, collapse = '')),og, 'xlogit', na.rm = T)
r= c(r, confint(r))
r = data.table(t(r))
setnames(r, c('result', 'lower', 'upper'))
})
r29.2 = rbindlist(r29.2)[, levels := as.character(levs)]
setorder(r29.2, levels)
expect_equivalent(r29.1, r29.2)
#more NA stuff
r30.1 = fake_sur[, mean(bin, na.rm = T), bin_by]
r30.2 = fake_sur[, smean(bin), bin_by]
r30.3 = svyby(~bin, ~bin_by, og, svymean) #NAs in by are ignored in survey
r30.4 = svymean(~bin, subset(og, is.na(bin_by)), na.rm = T)
expect_equivalent(r30.1, r30.2)
r31.1 = fake_sur[, sum(bin, na.rm = T), bin_by]
r31.2 = fake_sur[, stotal(bin), bin_by]
expect_equivalent(r31.1, r31.2)
#converting surveys when the dataset has been prefiltered
og_sub = subset(og, logi == T)
og_dt1 = dtsurvey(og_sub$variables, psu = 'psu', strata = 'strata', weight = 'weight', nest = T)
og_dt = as.dtsurvey(og_sub)
r32.1 = svymean(~num, og_sub)
r32.1 = data.table(result = r32.1, se = SE(r32.1))
r32.2 =og_dt[, smean(num, var_type = 'se')]
expect_equivalent(r32.1, r32.2)
#does sort order effect things
fs1 = copy(fake_sur)
fs2 = copy(fake_sur)
fs2 = fs2[sample(1:nrow(fs2), nrow(fs2), replace = F)]
expect_equivalent(fs1[, smean(num), keyby = fact], fs2[, smean(num), keyby = fact])
#0s in the weights
fake2 = copy(fake)
fake2[logi == T, weight :=0]
fake_sur2 = dtsurvey(fake2, psu = 'psu', strata = 'strata', weight = 'weight', nest = T)
r33.1 = fake_sur2[, smean(fact_na), logi]
r33.2 = fake_sur[, smean(fact_na), logi]
expect_true(r33.1[logi == TRUE, all(is.na(V1))])
expect_equivalent(r33.1[logi == FALSE], r33.2[logi == FALSE])
PHSKC-APDE/dtsurvey documentation built on March 6, 2024, 2:14 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
library('tinytest')
library('survey')
library('data.table')
library('dtsurvey')
#dumb survey
set.seed(98112)
size = 101
fake = data.table(num = runif(size),
fact = as.factor(sample(letters[1:5], size, TRUE)),
logi = sample(c(TRUE, FALSE), size, TRUE),
weight = 1,
strata = sample(1:10, size, TRUE),
byvar = sample(1:2, size, TRUE),
byna = sample(c(NA, 1, 2), size , TRUE))
fake[, psu := as.numeric(paste0(strata, sample(1:5, size, TRUE)))]
fake[, num_na := num]
fake[sample(seq_len(size), 20), num_na := NA]
fake[, fact_na := fact]
fake[sample(seq_len(size), 20), fact_na := NA]
levs = unique(fake[, fact])
fake[, paste0('fct_', levs) := lapply(levs, function(x) as.numeric(fact == x))]
fake[, paste0('fct_na_', levs) := lapply(levs, function(x) as.numeric(fact_na == x))]
fake[, bin := sample(0:1, size, T)]
fake[bin == 0, bin := NA]
fake[, bin_by := bin]
fake[, bin_fact := as.factor(bin)]
fake[!is.na(bin), bin2 := sample(0:1, .N, T)]
setorder(fake, byvar)
#make sure the survey package will take it
expect_silent(og <- survey::svydesign(~psu, strata = ~strata, data = fake, weights = ~weight), info = 'Fake data works with survey package')
fake_sur = dtsurvey(fake, psu = 'psu', strata = 'strata', weight = 'weight', nest = T)
#confirm that a mean, no bys or NAs are equal
r1.1 = svymean(~num, og)
r1.2 = fake_sur[, smean(num, var_type = 'se', ids = `_id`)]
expect_equal(as.numeric(r1.1), r1.2$result, info = 'Simple survey means are equal')
expect_equal(as.numeric(SE(r1.1)), r1.2$se, info = 'Simple survey ses are equal')
#this time with bys
r2.1 = svyby(~num, ~byvar, og, svymean)
r2.2 = fake_sur[, smean(num, var_type = 'se', ids = `_id`), by = byvar]
setorder(r2.2, byvar)
expect_equal(r2.1$num, r2.2$result, info = 'Simple survey means are equal, with one byvar')
expect_equal(as.numeric(SE(r2.1)), r2.2$se, info = 'Simple survey ses are equal, with one byvar')
#simple, with NA
r3.1 = svymean(~num_na, og, na.rm = T)
r3.2 = fake_sur[, smean(num_na, var_type = 'se', na.rm = TRUE, ids = `_id`)]
expect_equal(as.numeric(r3.1), r3.2$result, info = 'Simple survey means are equal, with some NAs')
expect_equal(as.numeric(SE(r3.1)), r3.2$se, info = 'Simple survey ses are equal, with some NAs')
#simple, with NA and by
r4.1 = svyby(~num_na, ~byvar, og, svymean, na.rm = TRUE)
r4.2 = fake_sur[, smean(num_na, var_type = 'se', ids = `_id`), by = byvar]
setorder(r4.2, byvar)
expect_equal(r4.1$num_na, r4.2$result, info = 'Simple survey means are equal, with one byvar and some NAs')
expect_equal(as.numeric(SE(r4.1)), r4.2$se, info = 'Simple survey ses are equal, with one byvar and some NAs')
#try assignments
expect_silent(r5 <- fake_sur[, blah := smean(num, ids = `_id`, var_type = NULL)], info = 'Simple survey mean assignment works')
expect_equal(r5[1, blah], as.numeric(svymean(~num, og, na.rm = T)))
expect_silent(r6 <- fake_sur[, blah := smean(num_na, ids = `_id`, na.rm = T, var_type = NULL), by = byvar], info = ' Simple survey mean assignment with by')
expect_equal(sort(unique(r6[, blah])), svyby(~num_na, ~byvar, og, svymean, na.rm = T)$num_na)
#means, se, degrees of freedom
expect_silent(r7.1 <- fake_sur[, smean(num, ids = `_id`, var_type = c('se', 'ci'), use_df = FALSE)])
r7.2 <- svymean(~num, og)
expect_equal(unname(unlist(r7.1)), unname(c(coef(r7.2), SE(r7.2), confint(r7.2))))
#Test alternative ways of calculating cis, first without by vars
r13.1 <- fake_sur[, smean(logi, ids = `_id`, var_type = 'ci', ci_method = 'xlogit')]
r13.2 <- fake_sur[, smean(logi, ids = `_id`, var_type = 'ci', ci_method = 'beta')]
r13.3 <- confint(svyciprop(~logi, og, method = 'xlogit', na.rm = T))
r13.4 <- confint(svyciprop(~logi, og, method = 'beta', na.rm = T))
expect_equal(unname(unlist(r13.1)[2:3]), as.vector(unname(r13.3)))
#alternative ci methods, with bys
r14.1 <- fake_sur[, smean(logi, ids = `_id`, var_type = 'ci', ci_method = 'xlogit'), byvar]
r14.2 <- fake_sur[, smean(logi, ids = `_id`, var_type = 'ci', ci_method = 'beta'), byvar]
r14.3 <- svyby(~logi, ~byvar, og, svyciprop, vartype = 'ci', method = 'xlogit', na.rm = T)
r14.4 <- svyby(~logi, ~byvar, og, svyciprop, vartype = 'ci', method = 'beta', na.rm = T)
expect_equal(unname(as.matrix(r14.1)), unname(as.matrix(r14.3)))
expect_equal(unname(as.matrix(r14.2)), unname(as.matrix(r14.4)))
#alternative ci methods, with filtering
r15.1 <- fake_sur[fact != 'a', smean(logi, ids = `_id`, var_type = 'ci', ci_method = 'xlogit'), byvar]
r15.2 <- fake_sur[fact != 'a', smean(logi, ids = `_id`, var_type = 'ci', ci_method = 'beta'), byvar]
r15.3 <- svyby(~logi, ~byvar, subset(og, fact != 'a'), svyciprop, vartype = 'ci', method = 'xlogit', na.rm = T)
r15.4 <- svyby(~logi, ~byvar, subset(og,fact != 'a'), svyciprop, vartype = 'ci', method = 'beta', na.rm = T)
expect_equal(unname(as.matrix(r15.1)), unname(as.matrix(r15.3)))
expect_equal(unname(as.matrix(r15.2)), unname(as.matrix(r15.4)))
#factors
r16.1 <- fake_sur[, smean(fact, ids = `_id`, var_type = 'se')]
r16.2 <- svymean(~fact, og, vartype = c('se'))
expect_equal(r16.1[, result], unname(coef(r16.2)))
expect_equal(r16.1[, se], unname(SE(r16.2)))
#factors, assign-- factor assignment is too annoying to get it to work. It probably needs to be wrapped in something else
# expect_silent(fake_sur[, blah_fact := smean(fact, ids = `_id`)])
# expect_true(all(unique(fake_sur[, blah_fact])[[1]] == fake_sur[, smean(fact, ids = `_id`)]))
#factors, by
r17.1 <- fake_sur[, smean(fact, ids = `_id`, var_type = c('se', 'ci'), use_df = FALSE), byvar]
r17.2 <- as.data.table(svyby(~fact, ~byvar, og, svymean))
r17.2 <- melt(r17.2, id.vars = 'byvar')
setorder(r17.2, byvar, variable)
expect_equal(r17.1[,result], r17.2[!substr(variable, 1,2) == 'se', value])
expect_equal(r17.1[,se], r17.2[substr(variable, 1,2) == 'se', value])
r17.3 = confint(svyby(~fact, ~byvar, og, svymean))
r17.1[, fact := rep(levels(fake_sur[, fact]),2)]
setorder(r17.1, fact, byvar)
expect_equal(unique(r17.1[,lower]), unname(r17.3[,1]))
expect_equal(unique(r17.1[,upper]), unname(r17.3[,2]))
#byna
r18.1 = fake_sur[, smean(num_na, ids = `_id`, var_type = 'se', use_df = FALSE), byna]
r18.2 = svyby(~num_na, ~byna, og, svymean, na.rm = T)
expect_false(length(r18.1[, result]) == length(r18.2[,2]), info = 'dtsurvey returns a row when a by var is NA, survey does not')
expect_equal((r18.1[!is.na(byna),result, keyby = byna][, result]), r18.2[,2], info = 'dtsurvey returns a row when a by var is NA, survey does not')
#multiby
r19.1 = fake_sur[, .(result = smean(num, ids = `_id`)), keyby = .(byvar, logi)] #todo, why does this return V1?
r19.2 = svyby(~num, ~byvar + logi, og, svymean, na.rm = T)
setorder(r19.1, logi, byvar)
expect_equal(r19.1[, result], r19.2[, 'num'])
#confirm that a total, no bys or NAs are equal
r20.1 = svytotal(~num, og)
r20.2 = fake_sur[, stotal(num, var_type = 'se', ids = `_id`)]
expect_equal(as.numeric(r20.1), r20.2$result, info = 'Simple rep survey totals are equal')
expect_equal(as.numeric(SE(r20.1)), r20.2$se, info = 'Simple rep totals ses are equal')
#this time with bys
r21.1 = svyby(~num, ~byvar, og, svytotal)
r21.2 = fake_sur[, stotal(num, var_type = 'se', ids = `_id`), by = byvar]
setorder(r21.2, byvar)
expect_equal(r21.1$num, r21.2$result, info = 'Simple rep survey totals are equal, with one byvar')
expect_equal(as.numeric(SE(r21.1)), r21.2$se, info = 'Simple rep survey ses are equal, with one byvar')
#simple, with NA
r22.1 = svytotal(~num_na, og, na.rm = T)
r22.2 = fake_sur[, stotal(num_na, var_type = 'se', na.rm = TRUE, ids = `_id`)]
expect_equal(as.numeric(r22.1), r22.2$result, info = 'Simple rep survey totals are equal, with some NAs')
expect_equal(as.numeric(SE(r22.1)), r22.2$se, info = 'Simple rep survey ses are equal, with some NAs')
#simple, with NA and by
r23.1 = svyby(~num_na, ~byvar, og, svytotal, na.rm = TRUE)
r23.2 = fake_sur[, stotal(num_na, var_type = 'se', ids = `_id`), by = byvar]
setorder(r23.2, byvar)
expect_equal(r23.1$num_na, r23.2$result, info = 'Simple rep survey totals are equal, with one byvar and some NAs')
expect_equal(as.numeric(SE(r23.1)), r23.2$se, info = 'Simple rep survey ses are equal, with one byvar and some NAs')
#factors, by
r24.1 <- fake_sur[, stotal(fact, ids = `_id`, var_type = c('se', 'ci'), use_df = FALSE), keyby = byvar]
r24.2 <- as.data.table(svyby(~fact, ~byvar, og, svytotal))
r24.2 <- melt(r24.2, id.vars = 'byvar')
setorder(r24.2, byvar, variable)
expect_equal(r24.1[, result], r24.2[!substr(variable, 1,2) == 'se', value])
expect_equal(r24.1[,se], r24.2[substr(variable, 1,2) == 'se', value])
r24.3 = confint(svyby(~fact, ~byvar, og, svytotal))
r24.1[, fact := rep(levels(fake_sur[, fact]),2)]
setorder(r24.1, fact, byvar)
expect_equal(unique(r24.1[,lower]), unname(r24.3[,1]))
expect_equal(unique(r24.1[,upper]), unname(r24.3[,2]))
#converting a survey object into a dtsurvey (see more below)
r25 = as.dtsurvey(og)
expect_equivalent(r25, dtsurvey(fake, psu = 'psu', strata = 'strata', weight = 'weight', nest = T))
r26 = survey::svydesign(~1, strata = ~strata, data = fake, weights = ~weight, nest = T)
expect_equivalent(as.dtsurvey(r26), dtsurvey(fake, psu = NULL, strata = 'strata', weight = 'weight', nest = T))
#ci method errors
#Proportion methods for factors
#x logit
r27.1 = fake_sur[, smean(fact, var_type = c('ci'), ci_method = 'xlogit')]
r27.1 = data.table(r27.1)
r27.2 = lapply(levs, function(x){
r = svyciprop(as.formula(paste0('~','fct_',x, collapse = '')),og, 'xlogit')
r= c(r, confint(r))
r = data.table(t(r))
setnames(r, c('result', 'lower', 'upper'))
})
r27.2 = rbindlist(r27.2)[, levels := as.character(levs)]
setorder(r27.2, levels)
expect_equivalent(r27.1, r27.2)
#beta
r28.1 = fake_sur[, smean(fact, var_type = c('ci'), ci_method = 'beta')]
r28.1 = data.table(r28.1)
r28.2 = lapply(levs, function(x){
r = svyciprop(as.formula(paste0('~','fct_',x, collapse = '')),og, 'beta')
r= c(r, confint(r))
r = data.table(t(r))
setnames(r, c('result', 'lower', 'upper'))
})
r28.2 = rbindlist(r28.2)[, levels := as.character(levs)]
setorder(r28.2, levels)
expect_equivalent(r28.1, r28.2)
#with NAs
r29.1 = fake_sur[, smean(fact_na, var_type = c('ci'), ci_method = 'xlogit')]
r29.1 = data.table(r29.1)
r29.2 = lapply(levs, function(x){
r = svyciprop(as.formula(paste0('~','fct_na_',x, collapse = '')),og, 'xlogit', na.rm = T)
r= c(r, confint(r))
r = data.table(t(r))
setnames(r, c('result', 'lower', 'upper'))
})
r29.2 = rbindlist(r29.2)[, levels := as.character(levs)]
setorder(r29.2, levels)
expect_equivalent(r29.1, r29.2)
#more NA stuff
r30.1 = fake_sur[, mean(bin, na.rm = T), bin_by]
r30.2 = fake_sur[, smean(bin), bin_by]
r30.3 = svyby(~bin, ~bin_by, og, svymean) #NAs in by are ignored in survey
r30.4 = svymean(~bin, subset(og, is.na(bin_by)), na.rm = T)
expect_equivalent(r30.1, r30.2)
r31.1 = fake_sur[, sum(bin, na.rm = T), bin_by]
r31.2 = fake_sur[, stotal(bin), bin_by]
expect_equivalent(r31.1, r31.2)
#converting surveys when the dataset has been prefiltered
og_sub = subset(og, logi == T)
og_dt1 = dtsurvey(og_sub$variables, psu = 'psu', strata = 'strata', weight = 'weight', nest = T)
og_dt = as.dtsurvey(og_sub)
r32.1 = svymean(~num, og_sub)
r32.1 = data.table(result = r32.1, se = SE(r32.1))
r32.2 =og_dt[, smean(num, var_type = 'se')]
expect_equivalent(r32.1, r32.2)
#does sort order effect things
fs1 = copy(fake_sur)
fs2 = copy(fake_sur)
fs2 = fs2[sample(1:nrow(fs2), nrow(fs2), replace = F)]
expect_equivalent(fs1[, smean(num), keyby = fact], fs2[, smean(num), keyby = fact])
#0s in the weights
fake2 = copy(fake)
fake2[logi == T, weight :=0]
fake_sur2 = dtsurvey(fake2, psu = 'psu', strata = 'strata', weight = 'weight', nest = T)
r33.1 = fake_sur2[, smean(fact_na), logi]
r33.2 = fake_sur[, smean(fact_na), logi]
expect_true(r33.1[logi == TRUE, all(is.na(V1))])
expect_equivalent(r33.1[logi == FALSE], r33.2[logi == FALSE])
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.