Parameter estimation in stochastic systems: some recent results and applications
Parameter estimation of sub-Gaussian stable distributions
In this paper, we present a parameter estimation method for sub-Gaussian stable distributions. Our algorithm has two phases: in the first phase, we calculate the average values of harmonic functions of observations and in the second phase, we conduct the main procedure of asymptotic maximum likelihood where those average values are used as inputs. This implies that the main procedure of our method does not depend on the sample size of observations. The main idea of our method lies in representing...
Parameter identification and synchronization of dynamical system by introducing an auxiliary subsystem.
Parametrization of multi-output autoregressive-regressive models for self-tuning control
Partially observed optimal controls of forward-backward doubly stochastic systems
The partially observed optimal control problem is considered for forward-backward doubly stochastic systems with controls entering into the diffusion and the observation. The maximum principle is proven for the partially observable optimal control problems. A probabilistic approach is used, and the adjoint processes are characterized as solutions of related forward-backward doubly stochastic differential equations in finite-dimensional spaces. Then, our theoretical result is applied to study a partially-observed...
Particle filter with adaptive sample size
The paper deals with the particle filter in state estimation of a discrete-time nonlinear non-Gaussian system. The goal of the paper is to design a sample size adaptation technique to guarantee a quality of a filtering estimate produced by the particle filter which is an approximation of the true filtering estimate. The quality is given by a difference between the approximate filtering estimate and the true filtering estimate. The estimate may be a point estimate or a probability density function...
Performance analysis of least squares algorithm for multivariable stochastic systems
In this paper, we consider the parameter estimation problem for the multivariable system. A recursive least squares algorithm is studied by minimizing the accumulative prediction error. By employing the stochastic Lyapunov function and the martingale estimate methods, we provide the weakest possible data conditions for convergence analysis. The upper bound of accumulative regret is also provided. Various simulation examples are given, and the results demonstrate that the convergence rate of the...
Periodical coefficient linear systems with random stationary input
Piecewise approximation and neural networks
The paper deals with the recently proposed autotracking piecewise cubic approximation (APCA) based on the discrete projective transformation, and neural networks (NN). The suggested new approach facilitates the analysis of data with complex dependence and relatively small errors. We introduce a new representation of polynomials that can provide different local approximation models. We demonstrate how APCA can be applied to especially noisy data thanks to NN and local estimations. On the other hand,...
Piecewise viscosity solutions for stochastic singular control.
Piece-wise optimal control in transport allocation problem
Pipelined language model construction for Polish speech recognition
The aim of works described in this article is to elaborate and experimentally evaluate a consistent method of Language Model (LM) construction for the sake of Polish speech recognition. In the proposed method we tried to take into account the features and specific problems experienced in practical applications of speech recognition in the Polish language, reach inflection, a loose word order and the tendency for short word deletion. The LM is created in five stages. Each successive stage takes the...
Polynomial asymptotic stability of damped stochastic differential equations.
Polynomial matrix solution to the discrete fixed-LAG smoothing problem
Predicting access to materialized methods by means of hidden Markov model
Pricing rules under asymmetric information
We consider an extension of the Kyle and Back's model [Back, Rev. Finance Stud.5 (1992) 387–409; Kyle, Econometrica35 (1985) 1315–1335], meaning a model for the market with a continuous time risky asset and asymmetrical information. There are three financial agents: the market maker, an insider trader (who knows a random variable V which will be revealed at final time) and a non informed agent. Here we assume that the non informed agent is strategic, namely he/she uses a utility function to...
Principal-agent approach to environmental improvements policies
Successful solution to any environmental problem implies working with Knightian uncertainty that explicitly deals with decision making under conditions of unstructured randomness. A 'wild' type of randomness that we will never discern due to its unstable properties makes the assignment of corresponding probabilities impossible. For that reason, the consideration of general economical factors within cost/benefit analysis must fail. So, instead of governmental intervention and a cup and trade scheme,...
Probabilistic and fuzzy arithmetic approaches for the treatment of uncertainties in the installation of torpedo piles.
Probabilistic mixture-based image modelling
During the last decade we have introduced probabilistic mixture models into image modelling area, which present highly atypical and extremely demanding applications for these models. This difficulty arises from the necessity to model tens thousands correlated data simultaneously and to reliably learn such unusually complex mixture models. Presented paper surveys these novel generative colour image models based on multivariate discrete, Gaussian or Bernoulli mixtures, respectively and demonstrates...
Problemas de control estocástico con información incompleta que admiten un proceso suficiente.
Se centra el estudio en los problemas de control estocástico con información incompleta de parámetro discreto.Se define para estos problemas un parámetro suficiente para el proceso básico y se demuestra que la clase de controles basados en éste es esencialmente completa.Como caso particular se estudia el modelo lineal normal y se ve la relación que existe entre el proceso suficiente definido para este modelo y el filtro de Kalman.