Method and system for identifying regeneration points in a Markov chain Monte Carlo simulation
AssigneeCarnegie Mellon University
InventorsJoseph B. Kadane, Anthony E. Brockwell
Life of this patent
Estimating the term…
Loading patent…
AssigneeCarnegie Mellon University
InventorsJoseph B. Kadane, Anthony E. Brockwell
Estimating the term…
The method of the present invention is to modify an initial target distribution it π by combining it with a point mass concentrated on an “artificial atom” α which is outside the state-space X. A Markov chain may then be constructed using any known technique (for example, using the Metropolis-Hastings Algorithm) with the new target distribution. For this chain, the state α is Harris-recurrent (i.e. with probability one, it occurs infinitely many times). By the Markov property, the times at which the new chain hits α are regeneration times. To recover an ergodic chain with limiting distribution π, it is sufficient simply to delete every occurrence of the state α from the new chain. The points immediately after the (deleted) occurrences of the state α are then regeneration times in a Markov chain with limiting distribution π.
A method of generating a sequence of random numbers using a computer, comprising:
A method of generating a sequence of random numbers using a computer, comprising:
A method of generating a sequence of random numbers in a distributed manner with a plurality of interconnected computers, said method comprising:
A memory device carrying a set of instructions which, when executed, performs a method on a computer comprising:
Dependencies are inferred from the claim text.