Reduced Rank Models for Multiple Time Series

Abstract

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. Biometrika Trust is collaborating with JSTOR to digitize, preserve and extend access to Biometrika. SUMMARY This paper is concerned with the investigation of reduced rank coefficient models for multiple time series. In particular, autoregressive processes which have a structure to their coefficient matrices similar to that of classical multivariate reduced rank regression are studied in detail. The estimation of parameters and associated asymptotic theory are derived. The exact correspondence between the reduced rank regression procedure for multiple autoregressive processes and the canonical analysis of Box & Tiao (1977) is briefly indicated. To illustrate the methods, U.S. hog data are considered.Some key word8: Canonical analysis; Canonical correlation; Multiple time series; Reduced rank regression.

Modelling Comovements of Economic Time Series: A Selective Survey

“…Optimal estimates of both the common features vectors and (partial) RRR coefficients are then obtained as described in Table 2 (see, inter alia, Velu et al, 1986;Ahn and Reinsel, 1988). …”

A Review of Forecasting Techniques for Large Data Sets

“…The model (2) was studied by Velu, Reinsel, and Wichern (1986). In (2), it is assumed that the true rank of the matrices α and β are identical and equal to r which is thus referred to as the rank of the system (2).…”

“…In (2), it is assumed that the true rank of the matrices α and β are identical and equal to r which is thus referred to as the rank of the system (2). Given the assumed system rank r, Velu, Reinsel, and Wichern (1986) suggested an estimation method for the parameters α and β that may be shown to be quasi-maximum likelihood (see also Reinsel and Velu (1998)). …”

Generalized Reduced Rank Regression

“…Discussion of these relations can be found in Anderson (1951), Izenman (1975), Tso (1981), Davies and Tso (1982), and Reinsel and Velu (1998). Reduced rank models with stationary variables have been analyzed by Velu, Reinsel, and Wichern (1986) and Velu and Reinsel (1987), Gudmundsson (1977) applied reduced rank regression to select linear combinations that best described the average variation in a vector of dependent variables. RRR is also used in the analysis of cointegrated variables.…”