scientific article; zbMATH DE number 3949563

From MaRDI portal
Publication:3719692





[For the entire collection see Zbl 0564.00016.] The linear discrete-time model \[ y_ t=\sum^{k}_{r=1}F_ r(\theta)y_{t-r}+\eta_ t;\quad \eta_ t=\sum^{\ell}_{r=0}C_ r(\theta)\epsilon_{t-r} \] \[ E(\epsilon_ t)=0,\quad E(\epsilon_ t\epsilon '\!_ t)=K(\theta),\quad E(\epsilon_ s\epsilon '\!_ t)=0,\quad s\neq t \] where the elements of the matrices \(F_ 1(\theta),...,F_ k(\theta)\), \(C_ 1(\theta),...,C_{\ell}(\theta)\) and K(\(\theta)\) are functions of a vector \(\theta\) of structural parameters of a continuous-time model, is very frequently applied in aggregate macroeconomic models and this is very closely related to the corresponding continuous-time stochastic models. This relationship is surveyed and discussed in some detail. In particular the estimation of parameters of stochastic differential equation systems such as \[ dy(t)=[A(\theta)y(t)+b(\theta)]dt+\zeta (dt) \] is discussed with reference to some empirical examples to show that these models can handle the prior restrictions imposed by economic theory much better than the discrete-time models.




Cited in
(50)








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 Q3719692)