credible.emc {EMC2} | R Documentation |
Posterior credible interval tests
Description
Modeled after t.test
, returns the credible interval of the parameter or test
and what proportion of the posterior distribution (or the difference in posterior distributions
in case of a two sample test) overlaps with mu.
For a one sample test provide x
and for two sample also provide y
.
Note that for comparisons within one model, we recommend using hypothesis()
if the priors
were well chosen.
Usage
## S3 method for class 'emc'
credible(
x,
x_name = NULL,
x_fun = NULL,
x_fun_name = "fun",
selection = "mu",
y = NULL,
y_name = NULL,
y_fun = NULL,
y_fun_name = "fun",
x_subject = NULL,
y_subject = NULL,
mu = 0,
alternative = c("less", "greater")[1],
probs = c(0.025, 0.5, 0.975),
digits = 2,
p_digits = 3,
print_table = TRUE,
...
)
credible(x, ...)
Arguments
x |
An emc object |
x_name |
A character string. Name of the parameter to be tested for |
x_fun |
Function applied to the MCMC chains to create variable to be tested. |
x_fun_name |
Name to give to quantity calculated by |
selection |
A character string designating parameter type (e.g. |
y |
A second emc object |
y_name |
A character string. Name of the parameter to be tested for |
y_fun |
Function applied to the MCMC chains to create variable to be tested. |
y_fun_name |
Name to give to quantity calculated by |
x_subject |
Integer or name selecting a subject |
y_subject |
Integer or name selecting a subject |
mu |
Numeric. |
alternative |
|
probs |
Vector defining quantiles to return. |
digits |
Integer, significant digits for estimates in printed results |
p_digits |
Integer, significant digits for probability in printed results |
print_table |
Boolean (defaults to |
... |
Additional optional arguments that can be passed to |
Value
Invisible results table with no rounding.
Examples
## Not run:
# Run a credible interval test (Bayesian ''t-test'')
# Here the full model is an emc object with the hypothesized effect
design_full <- design(data = forstmann,model=DDM,
formula =list(v~0+S,a~E, t0~1, s~1, Z~1, sv~1, SZ~1),
constants=c(s=log(1)))
full_model <- make_emc(forstmann, design_full)
full_model <- fit(full_model)
credible(full_model, x_name = "v")
# We can also compare between two sets of emc objects
# Now without a ~ E
design_null <- design(data = forstmann,model=DDM,
formula =list(v~0+S,a~1, t0~1, s~1, Z~1, sv~1, SZ~1),
constants=c(s=log(1)))
null_model <- make_emc(forstmann, design_null)
null_model <- fit(null_model)
credible(x = null_model, x_name = "a", y = full_model, y_name = "a")
# Or provide custom functions
credible(x = full_model, x_fun = function(d) d["a_Eaccuracy"] - d["a_Eneutral"])
## End(Not run)