initHMM: Initializing treeHMM with given parameters

In treeHMM: Tree Structured Hidden Markov Model

Description

Initializing treeHMM with given parameters

Usage

initHMM(States, Symbols, treemat, startProbs = NULL, transProbs = NULL,
  emissionProbs = NULL)

Arguments

States	A (2 * 1) vector with first element being discrete state value for the cases(or positive) and second element being discrete state value for the controls(or negative) for given treeHMM
Symbols	List containing (M * 1) vectors for discrete values of emissions(where "M" is the possible number of emissions) for each feature variable
treemat	Adjacent Symmetry Matrix that describes the topology of the tree
startProbs	(N * 1) vector containing starting probabilities for the states, where "N" is the possible number of states(Optional). Default is equally probable states
transProbs	(N * N) matrix containing transition probabilities for the states, where "N" is the possible number of states(Optional)
emissionProbs	List of (N * M) matrices containing emission probabilities for the states, for each feature variable(optional). Default is equally probable emissions

Value

List describing the parameters of treeHMM(pi, alpha, beta)

Examples

tmat = matrix(c(0,0,1,0,0,0,0,1,0,0,0,0,0,1,1,0,0,0,0,0,0,0,0,0,0),
               5,5, byrow= TRUE ) #for "X" (5 nodes) shaped tree
states = c("P","N") #"P" represent cases(or positive) and "N" represent controls(or negative)
hmmA = initHMM(states,list(c("L","R")), tmat) #one feature with two discrete levels "L" and "R"
hmmB = initHMM(states, list(c("X","Y")),tmat, c(0.5,0.5), matrix(c(0.7,0.3,0.3,0.7),2,2))

treeHMM documentation built on Dec. 16, 2019, 1:38 a.m.