IC 26529
INSTITUTO DE CALCULO REBECA CHEREP DE GUBER
Unidad Ejecutora - UE
artículos
Título:
Sparse estimation of Dynamic Principal Components for Forecasting of High Dimensional Time Series
Autor/es:
YOHAI; VICTOR J.; SMUCLER, EZEQUIEL; PEÑA, DANIEL
Revista:
International Journal of Forecasting
Editorial:
Elsevier
Referencias:
Lugar: Amsterdam; Año: 2021 vol. 37 p. 1498 - 1508
ISSN:
0169-2070
Resumen:
We present the sparse estimation of One Sided Dynamic Principal Components(ODPC) to forecast high dimensional time series. The forecast can be made di-rectly with the ODPC or using them as estimates of the factors in a GeneralizedDynamic Factor model. It is shown that a large reduction in the number of param-eter estimated for the ODPCs can be achieved without aecting their forecastingperformance.
