contribution of an exogenous variable
[contrib_x]=contrib(vari,mainf_proxy)
* vari = the exogenous variable, either a ts, a vector or such an object
* mainf_proxy = the nbterms first terms of the Infinite Moving average
* dropna = 'dropna' if the user wants to remove the NA values from the data (optional)
* contrib_x = the vector of contributions
load(GROCERDIR+'/macros/grocer/db/bdhenderic.dat'); rnet_mainf=mainf([1-0.1746071-0.0928556 ; 0.1746071],-0.6296264); rnet_contrib=contrib(delts(rnet),rnet_mainf) load(GROCERDIR+'/macros/grocer/db/bdhenderic.dat'); rnet_mainf=mainf([1-0.1746071-0.0928556 ; 0.1746071],-0.6296264,107);rnet_contrib=contrib(delts(rnet),rnet_mainf) // The two examples calculate the contribution of the interest rate to the delta of logarithm of m1 according to Hendry and Ericsson // model (estimated by grocer function hendryericsson(). In the first one, the moving average approximation has been calculated // only for the 100 first terms (default of mainf); Therefore contrib has given the message "WARNING:# of terms in the response // function are lower than the # of observations". In the second example, this number of terms has been adjusted to the length // of the ts and the warning message no longer occurs. | ![]() | ![]() |