/* This program is intended to demonstrate the effects of a pulse change in x(t) on the quality spread regression (quarterly data). The model we will examine is the ARDL model Pip(t) = 0.005595 + 0.33Pip(t-1) - 0.0243spdiff(t) - 0.0185spdiff(t-1). */ /* Here we examine the effect of keeping x(t) (here spdiff(t)) at zero for t = -20, -10, ... -1 and then at time t = 0 make x(t) = 1 and then x(t) = 0 again thereafter (i.e. a temporary change in xt). This will help us plot the lagged effects on the quality spread equation. */ Options Nodate; data dynamic; yt=0; yt1=0.008; xt=0; xt1=0; xt2=0; do t = -20 to 20; if t = 0 then xt = 1; if t = 1 then xt = 0; if t = 2 then xt1 = 0; yt = 0.005595 - 0.0243*xt - 0.0185*xt1 + 0.33*yt1; output; yt1=yt; xt2=xt1; xt1=xt; end; run; title 'Pulse change in spdiff(t) at time t = 0'; proc plot data=dynamic; plot xt*t; run; title1 'Effect of Pulse Change in spdiff(t) on Pip(t):'; title2 'Pip(t) = 0.005595 + 0.33Pip(t-1) - 0.0243spdiff(t) - 0.01855spdiff(t-1)'; proc plot data=dynamic; plot yt*t; run; proc print data=dynamic; run;