On The Peña–Box Model (Q4677044)
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 2169616
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | On The Peña–Box Model |
scientific article; zbMATH DE number 2169616 |
Statements
On The Peña–Box Model (English)
0 references
20 May 2005
0 references
canonical correlation analysis
0 references
factor model
0 references
multiple time series
0 references
0.84451336
0 references
0.7309218
0 references
0.72684497
0 references
0.7193525
0 references
0 references
0.71468306
0 references
0.70851403
0 references
A \(k\)-dimensional series \(Z_{t}\) is said to satisfy the Peña-Box model [see \textit{D. Peña} and \textit{G. E. P. Box}, J. Am. Stat. Assoc. 82, 836-843 (1987; Zbl 0623.62081)] with \(r\) factors if \(Z_{t}=HX_{t}+\varepsilon_{t}\), where \(X_{t}\) is an unobservable \(r\)-dimensional factor series \((k\geq r)\), \(H\) is an unknown constant matrix, \(\varepsilon_{t}\) is a \(k\)-dimensional white-noise series with covariance matrix \(\Sigma_{\varepsilon}\) of rank \(\Sigma_{\varepsilon}\geq0\), and \(X_{t}\) is fully independent of \(\varepsilon_{t}\). The authors study properties of the Peña-Box model under a stationarity assumption and propose a procedure, using a two-directional scalar component model to identify the model. The model and its invariant properties for the first-order non-stationary series are investigated. Model building procedures for both stationary series and first-order non-stationary series are presented. Simulation studies are used to illustrate the procedure and its finite sample behaviour. Two data sets, US hog data and wheat-price data, are re-analysed using the proposed procedure.
0 references