Simple conditions for convergence of sequential Monte Carlo genealogies with applications

06/30/2020
by   Suzie Brown, et al.
0

Sequential Monte Carlo algorithms are popular methods for approximating integrals in problems such as non-linear filtering and smoothing. Their performance depends strongly on the properties of an induced genealogical process. We present simple conditions under which the limiting process, as the number of particles grows, is a time-rescaled Kingman coalescent. We establish these conditions for standard sequential Monte Carlo with a broad class of low-variance resampling schemes, as well as for conditional sequential Monte Carlo with multinomial resampling.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset