The expected number of visits to a state before absorption satisfies the reward equationwith zero boundary data at absorbing states where counting stops.
For simple random walk on a finite connected loopless unweighted graph, start at and stop on first hitting . Ifthen and for the Graph Laplacian . The incoming-visit balance equation is for ; at the Laplacian value is because all its coordinates sum to zero. Thus is the voltage for unit current from to , and . In particular, the expected number of directed transitions before is .
For the chain with absorbing, equal-probability nearest-neighbour moves between , and a probability- self-loop at , starting from the expected absorption time is , the probability of visiting before is , and the expected number of visits to before absorption is .
Articles by others on the same topic
There are currently no matching articles.