Internal class for a binary SBM fit under MNAR sampling designs (double-standard, block-node, block-dyad).

It is not designed to be called directly by the user.

Super classes

sbm::SBM -> sbm::SimpleSBM -> SimpleSBM_fit -> SimpleSBM_fit_noCov -> SimpleSBM_MNAR_noCov

Active bindings

vExpec

double: variational approximation of the expectation complete log-likelihood

Methods

Public methods

• SimpleSBM_MNAR_noCov$new()

• SimpleSBM_MNAR_noCov$update_parameters()

• SimpleSBM_MNAR_noCov$update_blocks()

• SimpleSBM_MNAR_noCov$clone()

Inherited methods

• sbm::SBM$print()
• sbm::SBM$rNetwork()
• sbm::SimpleSBM$plot()
• sbm::SimpleSBM$predict()
• sbm::SimpleSBM$rEdges()
• sbm::SimpleSBM$rMemberships()
• sbm::SimpleSBM$show()
• SimpleSBM_fit$doVEM()
• SimpleSBM_fit$get_state()
• SimpleSBM_fit$reorder()
• SimpleSBM_fit$set_state()

---

SimpleSBM_MNAR_noCov$new()

constructor for simpleSBM_fit for missSBM purpose

Usage

SimpleSBM_MNAR_noCov$new(networkData, clusterInit)

Arguments

networkData

a structure to store network under missing data condition: either a matrix possibly with NA, or a missSBM:::partlyObservedNetwork

clusterInit

Initial clustering: a vector with size ncol(adjacencyMatrix), providing a user-defined clustering with nbBlocks levels.

---

SimpleSBM_MNAR_noCov$update_parameters()

update parameters estimation (M-step)

Usage

SimpleSBM_MNAR_noCov$update_parameters(nu = NULL)

Arguments

nu

currently imputed values

---

SimpleSBM_MNAR_noCov$update_blocks()

update variational estimation of blocks (VE-step)

Usage

SimpleSBM_MNAR_noCov$update_blocks(log_lambda = 0)

Arguments

log_lambda

additional term sampling dependent used to de-bias estimation of tau

---

SimpleSBM_MNAR_noCov$clone()

The objects of this class are cloneable with this method.

Usage

SimpleSBM_MNAR_noCov$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.