Video

Monte Carlo algorithms have completely revolutionised statistical computation, allowing previously intractable models to be easily solved. In particular, Markov chain Monte Carlo (MCMC) algorithms have allowed for the use of Bayesian inference in a multitude of settings. One reason for their success is a strong theoretical foundation, allows us to validate the basic algorithms, provide numerous extensions and generalisations of the algorithms, clarify different algorithm options and tunings, and evaluate the results. This talk will present simple graphical examples, dynamically displayed using Java applets, to illustrate the workings of MCMC. It will emphasise the impact and importance of various theoretical MCMC issues including ergodicity, qualitative and quantitative convergence rates, optimal scalings, and adaptive MCMC. It will not assume any previous MCMC background, and is designed to be accessible to everyone.

The lecture was held in the Purdue Memorial Union East and West Faculty Lounges, with a reception following. 

The Alan B. Pritsker Scholars Distinguished Lecture Series is presented by the Purdue School of Industrial Engineering. It is named for the late Alan B. Pritsker, who made major contributions to industrial engineering, particularly in the area of computer simulation. The lecture series reinforces Pritsker's legacy at Purdue and in the field of industrial engineering. As a leader in simulation, Pritsker was pivotal in building the School of Industrial Engineering's reputation, helping it to become a frontrunner in research and thought leadership. Pritsker was on the Purdue faculty from 1970-98. In 1973 he co-founded Pritsker & Associates to use computer simulation technologies to solve problems in industry and government. In 1989 Pritsker Corporation was formed from the merger of Pritsker & Associates and FACTROL.