make_priors {CausalQueries}R Documentation

Make Priors


A flexible function to generate priors for a model.


  alphas = NA,
  distribution = NA,
  node = NA,
  label = NA,
  statement = NA,
  confound = NA,
  nodal_type = NA,
  param_names = NA,
  param_set = NA



A model created with make_model


Real positive numbers giving hyperparameters of the Dirichlet distribution


String (or list of strings) indicating a common prior distribution (uniform, jeffreys or certainty)


A string (or list of strings) indicating nodes for which priors are to be altered


A string. Label for nodal type indicating nodal types for which priors are to be altered


A causal query (or list of queries) that determines nodal types for which priors are to be altered


A confound named list that restricts nodal types for which priors are to be altered. Adjustments are limited to nodes in the named list.


A string. Label for nodal type indicating nodal types for which priors are to be altered


A string. The name of specific parameter in the form of, for example, 'X.1', 'Y.01'


A string. Indicates the name of the set of parameters to be modified (useful when setting confounds)


Seven arguments govern *which* parameters should be altered. The default is 'all' but this can be reduced by specifying

* label or nodal_type The label of a particular nodal type, written either in the form Y0000 or Y.Y0000

* node, which restricts for example to parameters associated with node 'X'

* statement, which restricts for example to nodal types that satisfy the statement 'Y[X=1] > Y[X=0]'

* confound, which restricts for example to nodal types that satisfy the statement 'Y[X=1] > Y[X=0]'

* param_set, which us useful when setting confound statements that produces several sets of parameters

* param_names, which restricts in specific parameters by naming them

Two arguments govern what values to apply:

* alphas is one or more non negative numbers and

* distribution indicates one of a common class: uniform, jeffreys, or 'certain'

Any arguments entered as lists or vectors of size > 1 should be of the same length as each other.


A vector indicating the hyperparameters of the prior distribution of the nodal types.

For instance confound = list(X = Y[X=1]> Y[X=0]) adjust parameters on X that are conditional on nodal types for Y.

See Also

Other priors: get_priors(), make_par_values_multiple(), make_par_values(), make_values_task_list(), set_priors()


# Pass all nodal types
model <- make_model("Y <- X")
make_priors(model, alphas = .4)
make_priors(model, distribution = "jeffreys")

# Passing by names of node, parameter set or label
model <- make_model('X -> M -> Y')
make_priors(model, param_name = "X.1", alphas = 2)
make_priors(model, node = 'X', alphas = 3)
make_priors(model, param_set = 'Y', alphas = 5)
make_priors(model, node = c('X', 'Y'), alphas = 3)
make_priors(model, param_set = c('X', 'Y'), alphas = 5)
make_priors(model, node = list('X', 'Y'), alphas = list(3, 6))
make_priors(model, param_set = list('X', 'Y'), alphas = list(4, 6))
make_priors(model, node = c('X', 'Y'), distribution = c('certainty', 'jeffreys'))
make_priors(model, param_set = c('X', 'Y'), distribution = c('jeffreys', 'certainty'))
make_priors(model, label = '01', alphas = 5)
make_priors(model, node = 'Y', label = '00', alphas = 2)
make_priors(model, node =c('M', 'Y'), label = '11', alphas = 4)

# Passing a causal statement
make_priors(model, statement = 'Y[M=1] > Y[M=0]', alphas = 3)
make_priors(model, statement = c('Y[M=1] > Y[M=0]', 'M[X=1]== M[X=0]'), alphas = c(3, 2))

# Passing a confound statement
model <- make_model('X->Y') %>%
 set_confound(list(X = 'Y[X=1] > Y[X=0]', X = 'Y[X=1] < Y[X=0]'))

            confound = list(X='Y[X=1] > Y[X=0]',
                            X='Y[X=1] < Y[X=0]'),
            alphas = c(3, 6))

make_priors(model, confound= list(X='Y[X=1] > Y[X=0]'), alphas = 4)
make_priors(model, param_set='X_1', alphas = 5)
make_priors(model, param_names='X_2.1', alphas = .75)

make_model('X -> Y') %>%
  set_confound(list(X = 'Y[X=1]>Y[X=0]'))%>%
  make_priors(statement = 'X[]==1',
              confound = list(X = 'Y[X=1]>Y[X=0]', X = 'Y[X=1]<Y[X=0]'),
              alphas = c(2, .5))

[Package CausalQueries version 0.0.3 Index]