SER Y=MS(X,W1,W2,...,Wm) TO <data> computes linear combinations of lagged values of X as Y[t]=W1*X[t-m+1]+W2*X[t-m+1]+...+Wm*X[t] (without rescaling weights W1,...,Wm). By giving one of the weights with a terminating 'T' (say WkT), Y[t]=W1*X[t-k+1]+W2*X[t-k+2]+...+Wk*X[t]+...+Wm*X[t-k+m] is computed. For example, SER Y=MS(X,1,-1,1T,-1) TO <data> computes Y[t]=X[t-2]-X[t-1]+X[t]-X[t+1] S = More information on SER operations