tests/testthat/test_check_inputs_adjustedcif.r
In RobinDenz1/adjustedCurves: Confounder-Adjusted Survival Curves and Cumulative Incidence Functions
suppressMessages(requireNamespace("survival"))
set.seed(42)
sim_dat <- readRDS(system.file("testdata",
"d_sim_crisk_n_20.Rds",
package="adjustedCurves"))
sim_dat$group <- as.factor(sim_dat$group)
test_that("variable not in data", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="grouup",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"grouup is not a valid column name in 'data'.")
})
test_that("variable has wrong type", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="x1",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("The column in 'data' specified by 'variable' ",
"needs to be a factor variable."), fixed=TRUE)
})
test_that("variable has really wrong type", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable=1,
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("Arguments 'variable', 'ev_time', 'event' and ",
"'method' must be character strings, specifying ",
"variables in 'data'."), fixed=TRUE)
})
test_that("variable has wrong type with matching", {
sim_dat_err <- sim_dat
sim_dat_err$group <- as.character(sim_dat_err$group)
expect_error(check_inputs_adjustedcif(data=sim_dat_err,
variable="group",
ev_time="time",
event="event",
method="matching",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("The column in 'data' specified by 'variable' needs ",
"to be a factor variable."),
fixed=TRUE)
})
test_that("variable has wrong length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable=c("x1", "x2"),
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("'variable' must be a character string of length 1 ",
"specifying the grouping variable in 'data'."),
fixed=TRUE)
})
test_that("non-standard evaluation in variable", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable=group,
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("'variable' must be a character string ",
"specifying a variable in 'data'."))
})
test_that("variable is a constant", {
sim_dat_err <- sim_dat
sim_dat_err$group <- factor(1)
expect_error(check_inputs_adjustedcif(data=sim_dat_err,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"There have to be at least two groups in 'variable'.",
fixed=TRUE)
})
test_that("categorical variable not allowed", {
sim_dat_err <- sim_dat
sim_dat_err$group <- c(rep(0, 10), rep(1, 5), rep(2, 5))
sim_dat_err$group <- as.factor(sim_dat_err$group)
expect_error(check_inputs_adjustedcif(data=sim_dat_err,
variable="group",
ev_time="time",
event="event",
method="matching",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"Categorical treatments are currently not supported for method='matching'.")
})
test_that("ev_time not in data", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time2",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("time2 is not a valid column name in 'data'."))
})
test_that("ev_time wrong format", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="group",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"The column in 'data' specified by 'ev_time' must be numeric.")
})
sim_dat$time2 <- sim_dat$time
sim_dat$time2[1] <- -1
test_that("ev_time includes negative values", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time2",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"All values in the 'ev_time' variable must be >= 0.")
})
test_that("ev_time wrong length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time=c("time", "time"),
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("'ev_time' must be a character string of length 1 ",
"specifying the time until an event or ",
"censoring occured."))
})
test_that("non-standard evaluation in ev_time", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time=event_time,
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("'ev_time' must be a character string specifying ",
"a variable in 'data'."))
})
test_that("event not in data", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event2",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"event2 is not a valid column name in 'data'.")
})
test_that("event wrong format", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="group",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"The column in 'data' specified by 'event' must be numeric.")
})
sim_dat$event2 <- sample(c(0, 1), size=nrow(sim_dat), replace=TRUE)
test_that("event only includes two values", {
expect_warning(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event2",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("It is recommended to use the 'adjustedsurv' function",
" when the 'event' variable is binary."))
})
test_that("event wrong length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event=c("event", "event"),
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("'event' must be a character string of length 1 ",
"specifying the numeric event indicator."))
})
test_that("non-standard evaluation in event", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event=status,
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("'event' must be a character string ",
"specifying a variable in 'data'."))
})
test_that("cause has wrong type", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause="1"),
paste0("'cause' must be a number specifying the cause of ",
"interest in the column specified with 'event'."))
})
test_that("more than one cause supplied", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=c(1, 2)),
"'cause' must be of length = 1.")
})
test_that("method undefined", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="km3",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("Method 'km3' is undefined. See documentation ",
"for details on available methods."))
})
test_that("wrong method length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method=c("km", "km"),
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("'method' must be a single character string. Using ",
"multiple methods in one call is currently ",
"not supported."))
})
test_that("wrong conf_int type", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=1,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("'conf_int' must be either TRUE or FALSE."))
})
test_that("wrong conf_int length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=c(TRUE, TRUE),
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("'conf_int' must be either TRUE or FALSE, not a vector."))
})
test_that("wrong conf_level type", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level="0.95",
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"'conf_level' must be a number < 1 and > 0.")
})
test_that("conf_level not valid", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=-0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"'conf_level' must be a number < 1 and > 0.")
})
test_that("wrong conf_level length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=c(0.95, 0.95),
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"'conf_level' must be a number < 1 and > 0, not a vector.")
})
test_that("wrong bootstrap type", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=1,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"'bootstrap' must be either TRUE or FALSE.")
})
test_that("wrong bootstrap length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=c(TRUE, TRUE),
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"'bootstrap' must be either TRUE or FALSE, not a vector.")
})
test_that("wrong n_boot type", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=-2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"'n_boot' must be a positive integer > 1.")
})
test_that("wrong n_boot length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=c(2, 2),
na.action="na.omit",
clean_data=TRUE,
cause=1),
"'n_boot' must be a positive integer > 2, not a vector.")
})
test_that("wrong times", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times="string",
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"'times' must be a numeric vector or NULL.")
})
test_that("wrong na.action type", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=10,
na.action=1,
clean_data=TRUE,
cause=1),
"'na.action' must be a function or a single character string. See details.")
})
test_that("wrong na.action length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=10,
na.action=c("na.omit", "na.omit"),
clean_data=TRUE,
cause=1),
paste0("'na.action' must be a function or a character ",
"string, not a vector. See documentation."))
})
test_that("wrong clean_data type", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=10,
na.action="na.omit",
clean_data=1,
cause=1),
"'clean_data' must be either TRUE or FALSE.")
})
test_that("wrong clean_data length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=10,
na.action="na.omit",
clean_data=c(TRUE, TRUE),
cause=1),
"'clean_data' must be either TRUE or FALSE, not a vector.")
})
test_that("no extrapolation", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times=30,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("Values in 'times' must be smaller than ",
"max(data[,ev_time]). No extrapolation allowed."),
fixed=TRUE)
})
test_that("bootstrap only with treatment_model that can be updated", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="iptw_pseudo",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
treatment_model=runif(n=20),
cause=1),
paste0("'treatment_model' needs to be a model that can ",
"be refit or a formula object when using ",
"bootstrap=TRUE."))
})
test_that("no outcome_vars argument", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="direct_pseudo",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("Argument 'outcome_vars' must be specified when using ",
"method='direct_pseudo'. See documentation."))
})
test_that("wrong outcome_vars argument", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="direct_pseudo",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
outcome_vars=2,
cause=1),
paste0("'outcome_vars' should be a character vector of ",
"column names in 'data', used to model ",
"the outcome mechanism."))
})
test_that("wrong type_time argument", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="direct_pseudo",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
outcome_vars=c("x1"),
type_time="aha",
cause=1),
"'type_time' should be either 'factor', 'bs' or 'ns'.")
})
test_that("too many spline_df", {
expect_warning(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="direct_pseudo",
conf_int=FALSE,
conf_level=0.95,
times=c(0.01, 0.02),
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
outcome_vars=c("x1"),
type_time="bs",
spline_df=10,
cause=1),
"'spline_df' > len(times) might lead to problems.",
fixed=TRUE)
})
test_that("too many spline_df default", {
expect_warning(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="direct_pseudo",
conf_int=FALSE,
conf_level=0.95,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
outcome_vars=c("x1"),
type_time="bs",
times=c(0.1, 0.2, 0.4),
cause=1),
paste0("'spline_df' > length(times) might lead to problems ",
"when type_time!='factor'."),
fixed=TRUE)
})
test_that("not enough time points", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="direct_pseudo",
conf_int=TRUE,
conf_level=0.95,
times=0.2,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
outcome_vars=c("x1"),
cause=1),
paste0("'geese' models require at least two distinct ",
"time points. Add more points in time to ",
"'times' and run again."))
})
test_that("weights in matching", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="matching",
conf_int=FALSE,
conf_level=0.95,
times=0.2,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
treatment_model=runif(20, min=10, max=20),
cause=1),
paste0("Propensity Scores > 1 or < 0 not allowed. ",
"Perhaps you supplied weights on accident?"))
})
test_that("ps_scores > 1 or < 0", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aiptw_pseudo",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
outcome_vars=c("x1"),
cause=1,
treatment_model=rep(2, nrow(sim_dat))),
paste0("Propensity Scores supplied using the 'treatment_model' ",
"argument must be smaller than 1 and bigger than 0."))
})
test_that("ps_scores wrong length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aiptw_pseudo",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
outcome_vars=c("x1"),
cause=1,
treatment_model=c(0.1, 0.2)),
paste0("The vector of propensity score supplied in the ",
"'treatment_model' argument must be of ",
"length nrow(data)."), fixed=TRUE)
})
test_that("no treatment_model in matching", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="matching",
conf_int=FALSE,
conf_level=0.95,
times=0.2,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("Argument 'treatment_model' must be specified ",
"when using method='matching'."))
})
test_that("no treatment_model with method='aiptw'", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aiptw",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
outcome_model=NULL,
cause=1),
paste0("Argument 'treatment_model' must be specified ",
"when using method='aiptw'."))
})
test_that("no 'treatment_model' with iptw", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="iptw",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("Argument 'treatment_model' must be defined when ",
"using method='iptw'. See documentation."))
})
test_that("wrong 'treatment_model' with iptw", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="iptw",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1,
treatment_model="haha"),
paste0("'treatment_model' must be a glm or multinom object. ",
"See documentation."))
})
test_that("no outcome_model with method='aiptw'", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aiptw",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
treatment_model=NULL,
cause=1),
"Argument 'outcome_model' must be specified when using method='aiptw'.")
})
test_that("no 'treatment_model' with iptw_pseudo", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="iptw_pseudo",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("Argument 'treatment_model' must be specified when ",
"using method='iptw_pseudo'. See documentation."))
})
test_that("no outcome_model with method='direct'", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="direct",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"Argument 'outcome_model' must be specified when using method='direct'.")
})
test_that("wrong treatment_model with aiptw_pseudo", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aiptw_pseudo",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1,
outcome_vars=c("x1", "x2", "x4"),
treatment_model="wrong_treat_mod"),
paste0("Argument 'treatment_model' must be one of: 'glm', ",
"'multinom', 'mira' or a numeric vector of ",
"propensity scores."))
})
outcome_model <- list(cause=2)
class(outcome_model) <- "FGR"
test_that("cause is not the same as the cause in FGR", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="direct",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
outcome_model=outcome_model,
cause=1),
paste0("The FGR model needs to be fit with the same ",
"'cause' as specified in the 'cause' argument."))
})
test_that("conf_int with non-supported method", {
expect_warning(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="matching",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
treatment_model=NULL,
cause=1),
paste0("Asymptotic or exact variance calculations are ",
"currently not available for method='matching'. ",
"Use bootstrap=TRUE to get bootstrap estimates."))
})
## multiple imputation stuff
sim_dat_na_time <- sim_dat_na_group <- sim_dat_na_event <- sim_dat_na_x1 <-
sim_dat
sim_dat_na_time$time[3] <- NA
sim_dat_na_group$group[3] <- NA
sim_dat_na_event$event[3] <- NA
sim_dat_na_x1$x1[3] <- NA
mids_na_time <- suppressWarnings(mice::mice(sim_dat_na_time, m=2,
printFlag=FALSE))
mids_na_group <- suppressWarnings(mice::mice(sim_dat_na_group, m=2,
printFlag=FALSE))
mids_na_event <- suppressWarnings(mice::mice(sim_dat_na_event, m=2,
printFlag=FALSE))
mids_na_x1 <- suppressWarnings(mice::mice(sim_dat_na_x1, m=2, printFlag=FALSE))
outcome_model <- list()
class(outcome_model) <- "CauseSpecificCox"
censoring_model <- list()
class(censoring_model) <- "coxph"
treatment_model <- list()
class(treatment_model) <- "glm"
test_that("wrong outcome_model with mira", {
expect_error(check_inputs_adjustedcif(data=mids_na_x1,
variable="group",
ev_time="time",
event="event",
method="direct",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
cause=1,
na.action="na.omit",
clean_data=TRUE,
outcome_model=outcome_model),
paste0("When using multiple imputation, mira objects need ",
"to be supplied to 'outcome_model' ",
"instead of single models. See documentation."))
})
test_that("wrong treatment_model with mira", {
expect_error(check_inputs_adjustedcif(data=mids_na_x1,
variable="group",
ev_time="time",
event="event",
method="iptw",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1,
treatment_model=treatment_model),
paste0("When using multiple imputation, ",
"mira objects or a formula ",
"need to be supplied to 'treatment_model' instead of ",
"single models. See documentation."))
})
test_that("wrong censoring_model with mira", {
expect_error(check_inputs_adjustedcif(data=mids_na_x1,
variable="group",
ev_time="time",
event="event",
method="aiptw",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
cause=1,
na.action="na.omit",
clean_data=TRUE,
censoring_model=censoring_model),
paste0("When using multiple imputation, mira objects need to ",
"be supplied to 'censoring_model' instead of ",
"single models. See documentation."))
})
RobinDenz1/adjustedCurves documentation built on Sept. 27, 2024, 7:04 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.
suppressMessages(requireNamespace("survival"))
set.seed(42)
sim_dat <- readRDS(system.file("testdata",
"d_sim_crisk_n_20.Rds",
package="adjustedCurves"))
sim_dat$group <- as.factor(sim_dat$group)
test_that("variable not in data", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="grouup",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"grouup is not a valid column name in 'data'.")
})
test_that("variable has wrong type", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="x1",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("The column in 'data' specified by 'variable' ",
"needs to be a factor variable."), fixed=TRUE)
})
test_that("variable has really wrong type", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable=1,
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("Arguments 'variable', 'ev_time', 'event' and ",
"'method' must be character strings, specifying ",
"variables in 'data'."), fixed=TRUE)
})
test_that("variable has wrong type with matching", {
sim_dat_err <- sim_dat
sim_dat_err$group <- as.character(sim_dat_err$group)
expect_error(check_inputs_adjustedcif(data=sim_dat_err,
variable="group",
ev_time="time",
event="event",
method="matching",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("The column in 'data' specified by 'variable' needs ",
"to be a factor variable."),
fixed=TRUE)
})
test_that("variable has wrong length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable=c("x1", "x2"),
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("'variable' must be a character string of length 1 ",
"specifying the grouping variable in 'data'."),
fixed=TRUE)
})
test_that("non-standard evaluation in variable", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable=group,
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("'variable' must be a character string ",
"specifying a variable in 'data'."))
})
test_that("variable is a constant", {
sim_dat_err <- sim_dat
sim_dat_err$group <- factor(1)
expect_error(check_inputs_adjustedcif(data=sim_dat_err,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"There have to be at least two groups in 'variable'.",
fixed=TRUE)
})
test_that("categorical variable not allowed", {
sim_dat_err <- sim_dat
sim_dat_err$group <- c(rep(0, 10), rep(1, 5), rep(2, 5))
sim_dat_err$group <- as.factor(sim_dat_err$group)
expect_error(check_inputs_adjustedcif(data=sim_dat_err,
variable="group",
ev_time="time",
event="event",
method="matching",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"Categorical treatments are currently not supported for method='matching'.")
})
test_that("ev_time not in data", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time2",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("time2 is not a valid column name in 'data'."))
})
test_that("ev_time wrong format", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="group",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"The column in 'data' specified by 'ev_time' must be numeric.")
})
sim_dat$time2 <- sim_dat$time
sim_dat$time2[1] <- -1
test_that("ev_time includes negative values", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time2",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"All values in the 'ev_time' variable must be >= 0.")
})
test_that("ev_time wrong length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time=c("time", "time"),
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("'ev_time' must be a character string of length 1 ",
"specifying the time until an event or ",
"censoring occured."))
})
test_that("non-standard evaluation in ev_time", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time=event_time,
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("'ev_time' must be a character string specifying ",
"a variable in 'data'."))
})
test_that("event not in data", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event2",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"event2 is not a valid column name in 'data'.")
})
test_that("event wrong format", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="group",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"The column in 'data' specified by 'event' must be numeric.")
})
sim_dat$event2 <- sample(c(0, 1), size=nrow(sim_dat), replace=TRUE)
test_that("event only includes two values", {
expect_warning(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event2",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("It is recommended to use the 'adjustedsurv' function",
" when the 'event' variable is binary."))
})
test_that("event wrong length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event=c("event", "event"),
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("'event' must be a character string of length 1 ",
"specifying the numeric event indicator."))
})
test_that("non-standard evaluation in event", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event=status,
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("'event' must be a character string ",
"specifying a variable in 'data'."))
})
test_that("cause has wrong type", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause="1"),
paste0("'cause' must be a number specifying the cause of ",
"interest in the column specified with 'event'."))
})
test_that("more than one cause supplied", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=c(1, 2)),
"'cause' must be of length = 1.")
})
test_that("method undefined", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="km3",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("Method 'km3' is undefined. See documentation ",
"for details on available methods."))
})
test_that("wrong method length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method=c("km", "km"),
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("'method' must be a single character string. Using ",
"multiple methods in one call is currently ",
"not supported."))
})
test_that("wrong conf_int type", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=1,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("'conf_int' must be either TRUE or FALSE."))
})
test_that("wrong conf_int length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=c(TRUE, TRUE),
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("'conf_int' must be either TRUE or FALSE, not a vector."))
})
test_that("wrong conf_level type", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level="0.95",
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"'conf_level' must be a number < 1 and > 0.")
})
test_that("conf_level not valid", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=-0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"'conf_level' must be a number < 1 and > 0.")
})
test_that("wrong conf_level length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=c(0.95, 0.95),
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"'conf_level' must be a number < 1 and > 0, not a vector.")
})
test_that("wrong bootstrap type", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=1,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"'bootstrap' must be either TRUE or FALSE.")
})
test_that("wrong bootstrap length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=c(TRUE, TRUE),
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"'bootstrap' must be either TRUE or FALSE, not a vector.")
})
test_that("wrong n_boot type", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=-2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"'n_boot' must be a positive integer > 1.")
})
test_that("wrong n_boot length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=c(2, 2),
na.action="na.omit",
clean_data=TRUE,
cause=1),
"'n_boot' must be a positive integer > 2, not a vector.")
})
test_that("wrong times", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times="string",
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"'times' must be a numeric vector or NULL.")
})
test_that("wrong na.action type", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=10,
na.action=1,
clean_data=TRUE,
cause=1),
"'na.action' must be a function or a single character string. See details.")
})
test_that("wrong na.action length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=10,
na.action=c("na.omit", "na.omit"),
clean_data=TRUE,
cause=1),
paste0("'na.action' must be a function or a character ",
"string, not a vector. See documentation."))
})
test_that("wrong clean_data type", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=10,
na.action="na.omit",
clean_data=1,
cause=1),
"'clean_data' must be either TRUE or FALSE.")
})
test_that("wrong clean_data length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=10,
na.action="na.omit",
clean_data=c(TRUE, TRUE),
cause=1),
"'clean_data' must be either TRUE or FALSE, not a vector.")
})
test_that("no extrapolation", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aalen_johansen",
conf_int=TRUE,
conf_level=0.95,
times=30,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("Values in 'times' must be smaller than ",
"max(data[,ev_time]). No extrapolation allowed."),
fixed=TRUE)
})
test_that("bootstrap only with treatment_model that can be updated", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="iptw_pseudo",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
treatment_model=runif(n=20),
cause=1),
paste0("'treatment_model' needs to be a model that can ",
"be refit or a formula object when using ",
"bootstrap=TRUE."))
})
test_that("no outcome_vars argument", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="direct_pseudo",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("Argument 'outcome_vars' must be specified when using ",
"method='direct_pseudo'. See documentation."))
})
test_that("wrong outcome_vars argument", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="direct_pseudo",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
outcome_vars=2,
cause=1),
paste0("'outcome_vars' should be a character vector of ",
"column names in 'data', used to model ",
"the outcome mechanism."))
})
test_that("wrong type_time argument", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="direct_pseudo",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
outcome_vars=c("x1"),
type_time="aha",
cause=1),
"'type_time' should be either 'factor', 'bs' or 'ns'.")
})
test_that("too many spline_df", {
expect_warning(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="direct_pseudo",
conf_int=FALSE,
conf_level=0.95,
times=c(0.01, 0.02),
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
outcome_vars=c("x1"),
type_time="bs",
spline_df=10,
cause=1),
"'spline_df' > len(times) might lead to problems.",
fixed=TRUE)
})
test_that("too many spline_df default", {
expect_warning(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="direct_pseudo",
conf_int=FALSE,
conf_level=0.95,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
outcome_vars=c("x1"),
type_time="bs",
times=c(0.1, 0.2, 0.4),
cause=1),
paste0("'spline_df' > length(times) might lead to problems ",
"when type_time!='factor'."),
fixed=TRUE)
})
test_that("not enough time points", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="direct_pseudo",
conf_int=TRUE,
conf_level=0.95,
times=0.2,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
outcome_vars=c("x1"),
cause=1),
paste0("'geese' models require at least two distinct ",
"time points. Add more points in time to ",
"'times' and run again."))
})
test_that("weights in matching", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="matching",
conf_int=FALSE,
conf_level=0.95,
times=0.2,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
treatment_model=runif(20, min=10, max=20),
cause=1),
paste0("Propensity Scores > 1 or < 0 not allowed. ",
"Perhaps you supplied weights on accident?"))
})
test_that("ps_scores > 1 or < 0", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aiptw_pseudo",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
outcome_vars=c("x1"),
cause=1,
treatment_model=rep(2, nrow(sim_dat))),
paste0("Propensity Scores supplied using the 'treatment_model' ",
"argument must be smaller than 1 and bigger than 0."))
})
test_that("ps_scores wrong length", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aiptw_pseudo",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
outcome_vars=c("x1"),
cause=1,
treatment_model=c(0.1, 0.2)),
paste0("The vector of propensity score supplied in the ",
"'treatment_model' argument must be of ",
"length nrow(data)."), fixed=TRUE)
})
test_that("no treatment_model in matching", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="matching",
conf_int=FALSE,
conf_level=0.95,
times=0.2,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("Argument 'treatment_model' must be specified ",
"when using method='matching'."))
})
test_that("no treatment_model with method='aiptw'", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aiptw",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
outcome_model=NULL,
cause=1),
paste0("Argument 'treatment_model' must be specified ",
"when using method='aiptw'."))
})
test_that("no 'treatment_model' with iptw", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="iptw",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("Argument 'treatment_model' must be defined when ",
"using method='iptw'. See documentation."))
})
test_that("wrong 'treatment_model' with iptw", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="iptw",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1,
treatment_model="haha"),
paste0("'treatment_model' must be a glm or multinom object. ",
"See documentation."))
})
test_that("no outcome_model with method='aiptw'", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aiptw",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
treatment_model=NULL,
cause=1),
"Argument 'outcome_model' must be specified when using method='aiptw'.")
})
test_that("no 'treatment_model' with iptw_pseudo", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="iptw_pseudo",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
paste0("Argument 'treatment_model' must be specified when ",
"using method='iptw_pseudo'. See documentation."))
})
test_that("no outcome_model with method='direct'", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="direct",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1),
"Argument 'outcome_model' must be specified when using method='direct'.")
})
test_that("wrong treatment_model with aiptw_pseudo", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="aiptw_pseudo",
conf_int=FALSE,
conf_level=0.95,
times=NULL,
bootstrap=TRUE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1,
outcome_vars=c("x1", "x2", "x4"),
treatment_model="wrong_treat_mod"),
paste0("Argument 'treatment_model' must be one of: 'glm', ",
"'multinom', 'mira' or a numeric vector of ",
"propensity scores."))
})
outcome_model <- list(cause=2)
class(outcome_model) <- "FGR"
test_that("cause is not the same as the cause in FGR", {
expect_error(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="direct",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
outcome_model=outcome_model,
cause=1),
paste0("The FGR model needs to be fit with the same ",
"'cause' as specified in the 'cause' argument."))
})
test_that("conf_int with non-supported method", {
expect_warning(check_inputs_adjustedcif(data=sim_dat,
variable="group",
ev_time="time",
event="event",
method="matching",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
treatment_model=NULL,
cause=1),
paste0("Asymptotic or exact variance calculations are ",
"currently not available for method='matching'. ",
"Use bootstrap=TRUE to get bootstrap estimates."))
})
## multiple imputation stuff
sim_dat_na_time <- sim_dat_na_group <- sim_dat_na_event <- sim_dat_na_x1 <-
sim_dat
sim_dat_na_time$time[3] <- NA
sim_dat_na_group$group[3] <- NA
sim_dat_na_event$event[3] <- NA
sim_dat_na_x1$x1[3] <- NA
mids_na_time <- suppressWarnings(mice::mice(sim_dat_na_time, m=2,
printFlag=FALSE))
mids_na_group <- suppressWarnings(mice::mice(sim_dat_na_group, m=2,
printFlag=FALSE))
mids_na_event <- suppressWarnings(mice::mice(sim_dat_na_event, m=2,
printFlag=FALSE))
mids_na_x1 <- suppressWarnings(mice::mice(sim_dat_na_x1, m=2, printFlag=FALSE))
outcome_model <- list()
class(outcome_model) <- "CauseSpecificCox"
censoring_model <- list()
class(censoring_model) <- "coxph"
treatment_model <- list()
class(treatment_model) <- "glm"
test_that("wrong outcome_model with mira", {
expect_error(check_inputs_adjustedcif(data=mids_na_x1,
variable="group",
ev_time="time",
event="event",
method="direct",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
cause=1,
na.action="na.omit",
clean_data=TRUE,
outcome_model=outcome_model),
paste0("When using multiple imputation, mira objects need ",
"to be supplied to 'outcome_model' ",
"instead of single models. See documentation."))
})
test_that("wrong treatment_model with mira", {
expect_error(check_inputs_adjustedcif(data=mids_na_x1,
variable="group",
ev_time="time",
event="event",
method="iptw",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
na.action="na.omit",
clean_data=TRUE,
cause=1,
treatment_model=treatment_model),
paste0("When using multiple imputation, ",
"mira objects or a formula ",
"need to be supplied to 'treatment_model' instead of ",
"single models. See documentation."))
})
test_that("wrong censoring_model with mira", {
expect_error(check_inputs_adjustedcif(data=mids_na_x1,
variable="group",
ev_time="time",
event="event",
method="aiptw",
conf_int=TRUE,
conf_level=0.95,
times=NULL,
bootstrap=FALSE,
n_boot=2,
cause=1,
na.action="na.omit",
clean_data=TRUE,
censoring_model=censoring_model),
paste0("When using multiple imputation, mira objects need to ",
"be supplied to 'censoring_model' instead of ",
"single models. See documentation."))
})
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.