An ARMA type probability density estimator
Properties of a probability density estimator having the rational form of an ARMA spectrum are investigated. Under various conditions on the underlying density's Fourier coefficients, the ARMA estimator is shown to have asymptotically smaller mean integrated squared error (MISE) then the best tapered Fourier series estimator. The most interesting cases are those in which the Fourier coefficients \(\phi_ j\) are asymptotic to \(Kj^{-\rho}\) as \(j\to \infty\), where \(\rho >.\) For example, when \(\rho =2\) the asymptotic MISE of a certain ARMA estimator is only about 63\% of that for the optimum series estimator. For a density f with support in [0,\(\pi\) ], the condition \(\rho =2\) occurs whenever \(f'(0+)\neq 0\), \(f'(\pi -)=0\) and f is square integrable.
- The ARMA method of approximating probability density functions
- Consistency of an autorsgressive density estimator
- Efficiency of a Kernel Density Estimator Under an Autoregressive Dependence Model
- Integrated consistency of smoothed probability density estimators for stationary sequences
- On the estimation of the probability density by trigonometric series
This page was built for publication: An ARMA type probability density estimator
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1103296)
