Exercises – Summer school on foundations and advances in stochastic filtering (FASF 2015) Course 1: Nonlinear system identification using sequential Monte Carlo methods
نویسندگان
چکیده
(b) Simulate the model (1) to produce T = 100 measurements y1:T . Based on these measurements compute the optimal (in the sense that it minimizes the mean square error) estimate of xt | y1:t for t = 1, . . . , T . Implement a bootstrap particle filter and compare to the optimal estimates. You can for example perform this comparison by plotting the root mean square estimate (RMSE) ε(N) as a function of the number of particles used in the particle filter (also plot the RMSE for the optimal estimator in the same figure). The RMSE is defined according to ε(N) , √√√√ 1 T T ∑
منابع مشابه
Inference and Filtering for Partially Observed Diffusion Processes via Sequential Monte Carlo
Diffusion processes observed partially or discretely, possibly with observation error, arise when constructing stochastic models in continuous time. The method of Sequential Monte Carlo provides an alternative to Markov Chain Monte Carlo methods, and can be effective in complex models at the cutting edge of scientific research. This paper introduces Sequential Monte Carlo approaches to inferenc...
متن کاملExercises – Doctoral school on nonlinear system identification Guest lectures: Nonlinear system identification using sequential Monte Carlo methods
In other words, find the probability density functions f(·) and g(·) in (2) corresponding to the model (1). (b) Simulate the model (1) to produce T = 100 measurements y1:T . Based on these measurements compute the optimal (in the sense that it minimizes the mean square error) estimate of xt | y1:t for t = 1, . . . , T . Implement a bootstrap particle filter and compare to the optimal estimates....
متن کاملHomework – Chalmers Machine Learning Summer School Module: Solving inference problems using sequential Monte Carlo
These problems constitute a part of the course module Solving inference problems using sequential Monte Carlo, which is a part of the Chalmers Machine Learning Summer School, given at Chalmers University of Technology, Gothenburg, Sweden in April 2015. You are expected to complete at least two out of the three assignments provided below. The solutions should be presented with calculations as we...
متن کاملNonlinear System Identification Using Particle Filters
Particle filters are computational methods opening up for systematic inference in nonlinear/non-Gaussian state space models. The particle filters constitute the most popular sequential Monte Carlo (SMC) methods. This is a relatively recent development and the aim here is to provide a brief exposition of these SMC methods and how they are key enabling algorithms in solving nonlinear system ident...
متن کاملSequential Monte Carlo methods for system identification
One of the key challenges in identifying nonlinear and possibly non-Gaussian state space models (SSMs) is the intractability of estimating the system state. Sequential Monte Carlo (SMC) methods, such as the particle filter (introduced more than two decades ago), provide numerical solutions to the nonlinear state estimation problems arising in SSMs. When combined with additional identification t...
متن کامل