catnet (version 1.00.0)

cnPredict-method: Prediction

Description

Predicts the 'not-available' elements in an incomplete sample.

Usage

cnPredict(object, data)

Arguments

object

a catNetwork

data

a data matrix or data.frame

Value

• An updated sample matrix

Details

If the data is a matrix of categorical values then the rows should represent the network nodes. Otherwise, the columns represent the nodes. Indices are integers that range from 1 to the number of categories of the corresponding node.

Prediction is made for those data nodes that are marked as not-available (NA) and is based on maximum probability criterion, thus the categorical value with the maximum probability is assigned.

Examples

cnet <- cnRandomCatnet(numnodes=10, maxParents=3, numCategories=3)
  ## generate a sample of size 2 and set nodes 8, 9 and 10 as not-available
  psamples <- matrix(1+rbinom(10*2, 2, 0.4), nrow=10)
  psamples[8, ] <- rep(NA, 2)
  psamples[9, ] <- rep(NA, 2)
  psamples[10, ] <- rep(NA, 2)
  ## make show sample rows are named after the network's nodes
  rownames(psamples) <- cnNodes(cnet)
  ## predict the values of nodes 8, 9 and 10
  newsamples <- cnPredict(object=cnet, data=psamples)