Biasing the Underlying Markov Process

• Detailed balance for trial/acceptance Markov process

• Often it happens that tij is small while c is large (or vice-versa)

  • even if product is of order unity, pij will be small because of min()

• The underlying TPM can be adjusted (biased) to enhance movement among states

  • bias can be removed in reverse trial probability, or acceptance
  • require in general

• • ideally, c will be unity (all trials accepted) even for a “large” change
    • rarely achieve this level of improvement
  • requires coordination of forward and reverse moves

Previous slide

Next slide

Back to first slide

View graphic version