Variance estimation in adaptive sequential Monte Carlo
نویسندگان
چکیده
Sequential Monte Carlo (SMC) methods represent a classical set of techniques to simulate sequence probability measures through simple selection/mutation mechanism. However, the associated selection functions and mutation kernels usually depend on tuning parameters that are first importance for efficiency algorithm. A standard way address this problem is apply adaptive sequential (ASMC) methods, which consist in exploiting information given by history sample tune parameters. This article concerned with variance estimation such ASMC methods. Specifically, we focus case where asymptotic coincides one “limiting” algorithm as defined Beskos et al. (Ann. Appl. Probab. 26 (2016) 1111–1146). We prove that, under natural assumptions, estimator introduced Lee Whiteley (Biometrika 105 (2018) 609–625) nonadaptive (i.e., SMC) also consistent To do this, introduce new expressed terms coalescent tree-based measures, explain its connection previous one. Our constructed tracing genealogy interacting particle system. The tools use connect study Markov chain SMC As such, they may give some insights when dealing complex genealogy-involved problems systems more general scenarios.
منابع مشابه
Neural Adaptive Sequential Monte Carlo
Sequential Monte Carlo (SMC), or particle filtering, is a popular class of methods for sampling from an intractable target distribution using a sequence of simpler intermediate distributions. Like other importance sampling-based methods, performance is critically dependent on the proposal distribution: a bad proposal can lead to arbitrarily inaccurate estimates of the target distribution. This ...
متن کاملAn Adaptive Sequential Monte Carlo Sampler
Sequential Monte Carlo (SMC) methods are not only a popular tool in the analysis of state–space models, but offer an alternative to Markov chain Monte Carlo (MCMC) in situations where Bayesian inference must proceed via simulation. This paper introduces a new SMC method that uses adaptive MCMC kernels for particle dynamics. The proposed algorithm features an online stochastic optimization proce...
متن کاملNeural Adaptive Sequential Monte Carlo Supplementary Material
This section reviews the basic SMC algorithm, beginning by recapitulating the setup described in the main text. Consider a probabilistic model comprising (possibly multi-dimensional) hidden and observed states z1:T and x1:T respectively, whose joint distribution factorizes as p(z1:T ,x1:T ) = p(z1)p(x1|z1) ∏T t=2 p(zt|z1:t−1)p(xt|z1:t,x1:t−1). This general form subsumes common statespace models...
متن کاملSequential Monte Carlo for rare event estimation
This paper discusses a novel strategy for simulating rare events and an associated Monte Carlo estimation of tail probabilities. Our method uses a system of interacting particles and exploits a FeynmanKac representation of that system to analyze their fluctuations. Our precise analysis of the variance of a standard multilevel splitting algorithm reveals an opportunity for improvement. This lead...
متن کاملStructural Estimation Using Sequential Monte Carlo Methods
0501, 0463) Structural Estimation Using Sequential Monte Carlo Methods
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Annals of Applied Probability
سال: 2021
ISSN: ['1050-5164', '2168-8737']
DOI: https://doi.org/10.1214/20-aap1611