scientific article; zbMATH DE number 3940528

From MaRDI portal
Publication:3711564





[For the entire collection see Zbl 0572.00017.] Stationary models for mixed time series and cross-section data are investigated. The models have the form of first- and second-order stochastic differential equations with coefficients varying randomly over cross-sectional units. E.g. the considered first-order model has the form \[ dY_{\omega}(t)/dt=\alpha_{\omega}Y_{\omega}(t)+\beta_{\omega}dX_ {\omega}(t)/dt, \] where \(\alpha_{\omega}\) and \(\beta_{\omega}\) are random variables that are mutually independent and independent of \(X_{\omega}(t)\), \(X_{\omega}(t)\) are unobservable independent processes with zero means and orthonormal increments \((-\infty <t<\infty\), \(\omega =1,2,...).\) The description of the models in continuous time enables to handle irregularly-spaced and cross-sectionally varying observation times. Problems of stationarity, distributional properties and parametric inference for the above models are solved in the paper. The behaviour of cross-sectionally aggregated data is also investigated.











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 Q3711564)