The Influence of Numerical and Observational Errors on the Likelihood of an ARMA Series (Q4956030)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 1452432
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | The Influence of Numerical and Observational Errors on the Likelihood of an ARMA Series |
scientific article; zbMATH DE number 1452432 |
Statements
The Influence of Numerical and Observational Errors on the Likelihood of an ARMA Series (English)
0 references
24 May 2000
0 references
time series
0 references
ARMA series
0 references
likelihood function
0 references
error analysis
0 references
0.7378380298614502
0 references
0.7343432903289795
0 references
Consider an invertible, stationary autoregressive moving-avarage (ARMA) series \[ y_t-\phi_1y{t-1}-\dots-\phi_py{t-p}=a_t-\theta_1a_{t-1}-\dots -\theta_qa_{t-q}, \] where \(a_t\) are independent \(N(0,\sigma^2)\) random variables while \(\sigma\), \(\phi_1,\dots,\phi_p\) and \(\theta_1,\dots,\theta_q\) are parameters, \(p\) and \(q\) are given. The authors investigate the influence of observations and round-off errors on the computation accuracy of the likelihood in a such time series context. They focused on the method proposed by \textit{M. M. Ali} [Biometrika 64, 535-545 (1977; Zbl 0377.62051)] for calculating the likelihood function and derived ``error analysis'' formulae which describe how these errors propagate through a large number of computational steps.
0 references