Metropolis-Hastings algorithm
Metropolis-Hastings algorithm
Metropolis-Hastings algorithm
Create successful ePaper yourself
Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.
• Initial i assumptions:<br />
Monte Carlo approach<br />
– Prior distribution is specified by function (continuous or discrete) that is<br />
easily evaluated (by computer):<br />
• If specify θ, then p(θ) is easily determined.<br />
– Likelihood function, p(D | θ), can be determined for any specified values of<br />
D and θ.<br />
– (Actually, all that is required is that the product of the prior and the<br />
likelihood be easily determined.)<br />
• Method produces an approximation of the posterior distribution, p(θ | D):<br />
– Provides large number of θ values sampled from the posterior distribution.<br />
– Can be used to estimate:<br />
• Mean, median, standard ddeviation of posterior.<br />
• Credible HDI regions.<br />
• Etc.<br />
• Example of Monte Carlo method.