| jensenSDiv {MethylIT.utils} | R Documentation |
Compute Jensen-Shannon Divergence of probqbility vectors p and q.
jensenSDiv(p, q, Pi = 0.5, logbase = 2)
p, q |
Probability vectors, sum(p_i) = 1 and sum(q_i) = 1. |
Pi |
Weight of the probability distribution p. The weight for q is: 1 - Pi. Default Pi = 0.5. |
logbase |
A positive number: the base with respect to which logarithms |
The Jensen–Shannon divergence is a method of measuring the similarity between two probability distributions. Here, the generalization given in reference [1] is used. Jensen–Shannon divergence is expressed in terms of Shannon entroppy. 0 < jensenSDiv(p, q) < 1, provided that the base 2 logarithm is used in the estimation of the Shannon entropies involved.
1. J. Lin, “Divergence Measures Based on the Shannon Entropy,” IEEE Trans. Inform. Theory, vol. 37, no. 1, pp. 145–151, 1991.
set.seed(123) counts = sample.int(10) prob.p = counts/sum(counts) counts = sample.int(12,10) prob.q = counts/sum(counts) jensenSDiv(prob.p, prob.q)