For Vector Autoregressive models, the problem of dimensionality, associated with an increasing dimension of the model, can affect the power of noncausality tests. In this paper, by a Monte Carlo study, we analyze the impact of high dimensionality on the power of noncausality test and we proposed a testing strategy that, under certain conditions, limit the negative effects of high dimensionality in the causality analysis.
Dimensionality problem in testing for non causality between time series. A partial solution / DI IORIO, Francesca; Triacca, U.. - (2004), pp. 911-918. (Intervento presentato al convegno COMPSTAT 2004).
Dimensionality problem in testing for non causality between time series. A partial solution
DI IORIO, FRANCESCA;
2004
Abstract
For Vector Autoregressive models, the problem of dimensionality, associated with an increasing dimension of the model, can affect the power of noncausality tests. In this paper, by a Monte Carlo study, we analyze the impact of high dimensionality on the power of noncausality test and we proposed a testing strategy that, under certain conditions, limit the negative effects of high dimensionality in the causality analysis.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.