inspector (version 1.0.3)

inspect_bfactor: Validate vectors of Bayes factors

Description

inspect_bfactor checks if an object is a numeric vector of valid Bayes factor values. This can be useful to validate inputs, intermediate calculations or outputs in user-defined functions.

Usage

inspect_bfactor(x, allow_nas = TRUE, warning_nas = TRUE)

Arguments

x

An arbitrary object.

allow_nas

Logical value. If TRUE then NA and NaN values in x are allowed. If FALSE, execution is stopped and an error message is thrown in case there are NA or NaN values in x.

warning_nas

Logical value. If TRUE then the presence of NA or NaN values in x generates a warning message. NA and NaN values pass silently otherwise (if allow_nas is TRUE).

Value

inspect_bfactor does not return any output. There are three possible outcomes:

• The call is silent if:

  • x is a numeric vector of valid Bayes factor values and there are no NA or NaN values in x.

  • x is a numeric vector of valid Bayes factor values, there are some NA or NaN values in x, allow_nas is set to TRUE and warning_nas is set to FALSE.

• An informative warning message is given if x is a numeric vector of valid Bayes factor values, there are some NA or NaN values in x and both allow_nas and warning_nas are set to TRUE.

• An informative error message is thrown and the execution is stopped if:

  • x is not a numeric vector of valid Bayes factor values.

  • x is a numeric vector of valid Bayes factor values, there are some in NA or NaN values in x and allow_nas is set to FALSE.

Details

inspect_bfactor conducts a series of tests to check if x is a numeric vector of valid Bayes factor values. Namely, inspect_bfactor checks if:

• x is NULL or empty.

• x is an atomic vector.

• x is numeric.

• x has NA or NaN values.

• The values of x are non-negative.

See Also

• inspect_bfactor_log to check if an object is a numeric vector of valid logarithmic Bayes factor values.

• bfactor_interpret for the interpretation of Bayes factors.

• inspect_bfactor_scale to check if an object is a valid Bayes factor interpretation scale.

Examples

# NOT RUN {
# Calls that pass silently:
x1 <- c(0, 0.5, 1, 10, 50, 100)
x2 <- c(NA, 0.5, 1, 10, 50, 100)
inspect_bfactor(x1)
inspect_bfactor(x2, warning_nas = FALSE)
inspect_bfactor(x2, allow_nas = TRUE, warning_nas = FALSE)

# Call that throws an informative warning message:
y <- c(0.1, 0.2, NA, 0.4, 0.5)
try(inspect_bfactor(y))
try(inspect_bfactor(y, warning_nas = TRUE))
try(inspect_bfactor(y, allow_nas = TRUE, warning_nas = TRUE))

# Calls that throw informative error messages:
z <- c(-0.9, 0, 0.1, 0.2, 0.3, 0.4, 0.5)
try(inspect_bfactor(z))
mylist <- list(
  NULL, TRUE, factor(.5), matrix(0.5),
  "0.5", list(0.5), NA, NaN, numeric(0), -0.5, -5
)
try(inspect_bfactor(mylist[[1]]))
try(inspect_bfactor(mylist[[2]]))
try(inspect_bfactor(mylist[[3]]))
try(inspect_bfactor(mylist[[4]]))
try(inspect_bfactor(mylist[[5]]))
try(inspect_bfactor(mylist[[6]]))
try(inspect_bfactor(mylist[[7]]))
try(inspect_bfactor(mylist[[8]]))
try(inspect_bfactor(mylist[[9]]))
try(inspect_bfactor(mylist[[10]]))
try(inspect_bfactor(mylist[[11]]))
# }