A0503
Title: Reduced-rank matrix autoregressive models: A medium N approach
Authors: Ivan Ricardo - Maastricht University (Netherlands) [presenting]
Abstract: Reduced-rank regressions are powerful tools used to identify co-movements within economic time series. However, this task becomes challenging when matrix-valued time series are observed, where each dimension may have a different co-movement structure. Reduced-rank regressions are proposed with a tensor structure for the coefficient matrix to provide new insights into co-movements within and between the dimensions of matrix-valued time series. Moreover, the co-movement structures are related to two commonly used reduced-rank models, namely the serial correlation common feature and the index model. Two empirical applications involving U.S. states and economic indicators for the Eurozone and North American countries illustrate how the new tools identify co-movements
