Parameter estimation in systems of differential equations, based on three-way data arrays and random time changes (Q4258926)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 1337310
Language Label Description Also known as
default for all languages
No label defined
    English
    Parameter estimation in systems of differential equations, based on three-way data arrays and random time changes
    scientific article; zbMATH DE number 1337310

      Statements

      0 references
      0 references
      15 September 1999
      0 references
      maximum likelihood estimation
      0 references
      Ornstein-Uhlenbeck processes
      0 references
      random time changes
      0 references
      estimation of differential equations parameters
      0 references
      Parameter estimation in systems of differential equations, based on three-way data arrays and random time changes (English)
      0 references
      A system of \(n_1\times n_2\) differential equations of the form \(d X_{ij}=f_{ij}(\theta,X_{ij}(t))dt\), \(X_{ij}(t_0)=x_{ij0}\), is considered, where \(\theta\) is the vector of the unknown parameters, and \(f_{ij}\) are strictly positive. The problem is to estimate \(\theta\) from the observations \(x_{ijl}=X_{ij}(T(t_l))\), where \(t_l\), \(l=0,\dots,n_3\), are some known moments of time, and \(T_{ij}(t)\) are stochastic ``time-change'' functions. In the proposed model \(T_{ij}=t+\int_0^t U_{ij}(s)ds\), where \(U_{ij}\) is a multidimensional Ornstein-Uhlenbeck process, i.e. the solution of the system of stochastic differential equations \[ dU_{ij}(t)=-\beta U_{ij}(t)dt+g(\beta)\sum_{u=1}^{n_2} b_{ju} \sum_{h=1}^{n_1} a_{ih} dW_{hu}(t), \] where \(a_{uv}\), \(b_{hi}\), \(\beta\) are supposed unknown. The likelihood function is evaluated and an iterative algorithm for the MLE estimator computation is proposed.
      0 references

      Identifiers