Causality-in-variance

[James_]

Do you know hot to run the "Hafner and Herwartz" causality-in-variance test, or else called LM-GARCH?

It' s not particularly clear to me if this is the right way to do it:

-model the time series with GARCH model

-get the normalized residual

- open the obtained norm residuals in an Unrestricted VAR, check the lag criteria and choose the "best" lag

-run the Granger Causality