scientific article; zbMATH DE number 1409845
The authors consider a continuous time regression model of the form \(\nu(t)=a^{\tau}g(t)+\eta(t)\), \(t\in R\), where \(g(t)=[g_1(t),\ldots, g_{n}(t)]^{\tau}\) is a known vector function such that the functions \(g_{i}(t)>0\), \(i=1,\ldots,n\), are square integrable and form a linearly independent set. The process of errors \(\eta(t)\), \(t\in R\), has a representation \(\eta(t)=G(\xi(t))\), where \(\xi(t)\) is a measurable m.s. continuous stationary Gaussian process with \(E\xi(t)=0\), \(E\xi^2(t)=1\) and \[ B(|t|)= \text{cov}(\xi(0),\xi(t))=L(|t|)/ |t|^{\alpha},\quad 0<\alpha<1, \] \(L(t)\) is a real slowly varying function, and \(G(t)\) is a real function such that \(EG^2(\xi(0))<\infty.\) The function \(B(|t|)\) has a spectral density \(f(|\lambda|)\) which is a decreasing function for \(|\lambda|\geq \lambda_0.\) The authors consider estimates of the vector parameter \(\alpha\) of the form \[ (1)\qquad {\hat\alpha}(T)=(V(T))^{-1}\int\limits_0^{T}g(t)\nu(t) dt, \] where \(V(T)=\int\limits_0^{T}g(t)^{\tau} dt.\) Consistency of the estimates (1) is investigated. For example, the authors, under some assumptions, prove that for \(0<\alpha<{1/m}\) the following holds: \[ \lim\limits_{T\to\infty}\left\|dT^{-1}V(T)D({\hat a}(T)) V(T)d(T)^{-1}L^{-m}(T)T^{n\alpha-2} C_{m}^{-2}-l\right\|=0, \] where \(m=rank G\) and \[ l=\lim_{T\to\infty} d(T)^{-1}\int_0^1 \int_0^1 g(Tu) g(Tu)^{\tau}|u-v|^{-m\alpha}dudv d(T)^{-1}, \] \[ L=\lim_{T\to\infty} d(T)^{-1}V_1(T)d(T)^{-1},\qquad D({\hat a}(T))=E({\hat a}(T)-a))({\hat a}(T)-a)^{\tau}. \] The authors find a representation of the asymptotic distribution of the process \(X(T)= R(T)^{-1}({\hat a}(T)-a)\) for some \(R(T)\).
- scientific article; zbMATH DE number 1524394
- Asymptotic properties of the LSE in a regression model with long-memory Gaussian and non-Gaussian stationary errors
- On estimation of a regression model with long-memory stationary errors
- scientific article; zbMATH DE number 1738344
- Semiparametric analysis of long-range dependence in nonlinear regression
- On estimation of a regression model with long-memory stationary errors
- On the rate of convergence to the normal law for LSE in multivariate continuous regression model with long-range dependence stationary errors.
- Asymptotic inference for LSE in multivariate continuous regression models with long-memory random fields.
- Moment estimator for an AR(1) model driven by a long memory Gaussian noise
- S-estimation in the linear regression model with long-memory error terms under trend
- Spectral regression for cointegrated time series with long-memory innovations
- scientific article; zbMATH DE number 3940529 (Why is no real title available?)
- scientific article; zbMATH DE number 4020263 (Why is no real title available?)
- On estimation of regression coefficients of long memory random fields observed on the arrays
- scientific article; zbMATH DE number 1524394 (Why is no real title available?)
- Periodogram estimates in nonlinear regression models with long-range dependent noise
- Consistency of the least squares estimator of the amplitudes and angular frequencies of a sum of harmonic oscillations in models with long-range dependence
- On the estimation of the regression coefficients of a continuous parameter process with stationary residual
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4940255)
