data temp(drop = t);
        format x date9.;
        do x = "01Jan2011"d to "30Jan2011"d;
                t = abs(round(rannor(0))) + abs(round(100*rannor(0)));
                y + t;
                output;
        end;
run;

proc arima data = Temp;
        identify var = y(1) nlag=20;
run;
        estimate q = 1 method=uls;
run;
        forecast lead=10 interval=day id=x out=temp_out;
run;
quit;

如果y为3年内按日计数的数据，现在用来按日预测，考虑时间季节性、节日等因素，如何破？