These codes perform analysis of the Granger causality between two time series Xt,Yt through subset ARX(p,q) models which have an irregular structure. Namely, they have sparse coefficients within maximum order lags p,q. Model identification is carried out with backward stepwise OLS regression with heteoskedastic consistent (HC) standard errors. The indicators of causality are F-statistics (on reduction of the residual variance) and Gain parameters (with T-statistics). Recently a Robust M-estimator version is also provided with demos for additive outliers and GARCH residuals.
Robust Analysis of the Causality in Subset ARX Models
Grillenzoni, Carlo
2022-01-01
Abstract
These codes perform analysis of the Granger causality between two time series Xt,Yt through subset ARX(p,q) models which have an irregular structure. Namely, they have sparse coefficients within maximum order lags p,q. Model identification is carried out with backward stepwise OLS regression with heteoskedastic consistent (HC) standard errors. The indicators of causality are F-statistics (on reduction of the residual variance) and Gain parameters (with T-statistics). Recently a Robust M-estimator version is also provided with demos for additive outliers and GARCH residuals.
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.
