Reprojecting Partially Observed Systems with Application to Interest Rate Diffusions

From MaRDI portal





Reprojection is a general-purpose technique for estimating the transition dynamics of the observed variables of an estimated nonlinear dynamic model with latent variables, which are usually due to partial observation of the state vector. The authors use reprojection as the third step of a procedure. The first: data are summarized by projection onto a Hermite series representation of the unconstrained transition density for observables. The second: system parameters are estimated by minimum chi-squared. The third: the constraints on dynamics implied by the nonlinear system are imposed by projecting a long simulation of the estimated system onto a Hermite series representation of the constrained transition density for observables. This technique is useful for diagnostic evaluation of continuous-time diffusion models for the short-term interest rate, and presents evidence on the importance of an intercept term in the diffusion function, and gives a new continuous-time model that permits feedbacks from the interest rate into the drift and diffusion functions of stochastic volatility.




Cited in
(37)








This page was built for publication: Reprojecting Partially Observed Systems with Application to Interest Rate Diffusions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3839571)