topic Re: VAR Granger Causality- Proc Varmax in SAS Forecasting and Econometrics https://communities.sas.com/t5/SAS-Forecasting-and-Econometrics/VAR-Granger-Causality-Proc-Varmax/m-p/172009#M1085 <HTML><HEAD></HEAD><BODY><P>Hi!</P><P></P><P>The Akaike (AIC) or the Schwarz-Bayesian (SBC) criterion can be used to determine the proper order of p. It is in the SAS output.</P></BODY></HTML> Thu, 17 Jul 2014 11:15:17 GMT user24feb 2014-07-17T11:15:17Z VAR Granger Causality- Proc Varmax https://communities.sas.com/t5/SAS-Forecasting-and-Econometrics/VAR-Granger-Causality-Proc-Varmax/m-p/172008#M1084 <HTML><HEAD></HEAD><BODY><P>Hi,</P><P>I am trying to do VAR Granger Causality using Proc Varmax by the below code:</P><P></P><PRE style="padding: 1.4em 0 0 12px; font-size: 12px; font-family: monospace; color: #333333;"> <P>&nbsp; /*--- Causality Testing ---*/</P> <P>proc varmax data=grunfeld;</P> <P>model y1-y3 = x1 x2 / p=1 noprint;</P> <P>causal group1=(x1) group2=(y1-y3); causal group1=(y3) group2=(y1 y2);</P> <P>run; </P> <P></P> <P></P> </PRE><P>My question, how can I choose how many lag or&nbsp; the value for p- p=????? (I think it is how many lag to test but how shall I choose it)?</P><P>My results changed significantly if I change the value for p</P></BODY></HTML> Wed, 09 Jul 2014 02:34:07 GMT https://communities.sas.com/t5/SAS-Forecasting-and-Econometrics/VAR-Granger-Causality-Proc-Varmax/m-p/172008#M1084 boas3ad 2014-07-09T02:34:07Z Re: VAR Granger Causality- Proc Varmax https://communities.sas.com/t5/SAS-Forecasting-and-Econometrics/VAR-Granger-Causality-Proc-Varmax/m-p/172009#M1085 <HTML><HEAD></HEAD><BODY><P>Hi!</P><P></P><P>The Akaike (AIC) or the Schwarz-Bayesian (SBC) criterion can be used to determine the proper order of p. It is in the SAS output.</P></BODY></HTML> Thu, 17 Jul 2014 11:15:17 GMT https://communities.sas.com/t5/SAS-Forecasting-and-Econometrics/VAR-Granger-Causality-Proc-Varmax/m-p/172009#M1085 user24feb 2014-07-17T11:15:17Z