transitionScale {VanillaICE} | R Documentation |
Accessor and replacement methods for scaling the transition probabilities
transitionScale(object) transitionScale(object) <- value
object |
An object of class HmmParameter |
value |
{A matrix of dimension STATES x STATES, where STATES is the number of hidden states.}
The probability of remaining in the same state, $P(S_t = S_{t+1})$
(the diagonal of the transition probability matrix) is a function of the
distance between SNPs. The probability of transitioning to some other
state is epsilon, where epsilon = 1 - $P(S_t = S_{t+1})$. The epsilon
is split among STATES-1 states. By default, the probability of
transitioning from an altered state back to the normal state is twice as
likely as the probability of transitioning between two altered states.
The weights for epsilon are provided in the transitionScale
matrix in objects of class HmmParameter
.
A matrix
R. Scharpf