Approximate Methods for State-Space Models
نویسندگان
چکیده
منابع مشابه
Approximate Methods for State-Space Models.
State-space models provide an important body of techniques for analyzing time-series, but their use requires estimating unobserved states. The optimal estimate of the state is its conditional expectation given the observation histories, and computing this expectation is hard when there are nonlinearities. Existing filtering methods, including sequential Monte Carlo, tend to be either inaccurate...
متن کاملApproximate inference for state-space models
This thesis is concerned with state estimation in partially observed diffusion processes with discrete time observations. This problem can be solved exactly in a Bayesian framework, up to a set of generally intractable stochastic partial differential equations. Numerous approximate inference methods exist to tackle the problem in a practical way. This thesis introduces a novel deterministic app...
متن کاملDeterministic approximate inference techniques for conditionally Gaussian state space models
We describe a novel deterministic approximate inference technique for conditionally Gaussian state space models, i.e. state space models where the latent state consists of both multinomial and Gaussian distributed variables. The method can be interpreted as a smoothing pass and iteration scheme symmetric to an assumed density filter. It improves upon previously proposed smoothing passes by not ...
متن کاملSymbolic Methods for the State Space Exploration of GSPN Models
Generalised Stochastic Petri Nets (GSPNs) suffer from the same problem as any other state-transition modelling technique: it is difficult to represent sufficient states so that general, real life systems can be analysed. In this paper we use symbolic techniques to perform state space exploration for unstructured GSPNs. We present an algorithm for finding an encoding function which attempts to m...
متن کاملImproved Monte Carlo Methods for State-Space Models
The recent availability of low cost powerful computational resources has led to the development of a plethora of Monte Carlo based inference mechanisms for use in complex statistical problems, many of which involve temporally evolving systems. Sequential Monte Carlo (SMC), which is based around Sequential Importance Sampling and Resampling, is a class of Monte Carlo algorithm that can be applie...
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: Journal of the American Statistical Association
سال: 2010
ISSN: 0162-1459,1537-274X
DOI: 10.1198/jasa.2009.tm08326