数模——时间序列与非参数统计习题答案

时间序列第1题 data a;input train@@; time=intnx('month','1Jan1993'd,_n_-1);format time monyy.; dif=dif(train); cards; 1196.8 1181.3 1222.6 1229.3 1221.5 1148.4 1250.2 1174.4 1234.5 1209.7 1206.5 1204.0 1234.1 1146.0 1304.9 1221.9 1244.1 1194.4 1281.5 1277.3 1238.9 1267.5 1200.9 1245.5 1249.9 1220.1 1267.4 1182.3 1221.7 1178.1 1261.6 1274.5 1196.4 1222.6 1174.7 1212.6 1215.0 1191.0 1179.0 1224.0 1183.0 1288.0 1274.0 1218.0 1263.0 1205.0 1210.0 1243.0 1266.0 1200.0 1306.0 1209.0 1248.0 1208.0 1231.0 1244.0 1296.0 1221.0 1287.0 1191.0 ; proc gplot;plot train*time=1 dif*time=2; symbol1 v=circle i=join c=black; symbol2 v=star i=join c=red; proc arima data=a;identify var=dif nlag=22 minic p=(0:5) q=(0:5); run;

模型为ARIMA(1,1) 加上命令proc arima data=a;identify var=dif nlag=22;estimate p=1 q=1 method=ml;

时间序列第2题 data b;input cons@@; time=intnx('year','1952',_n_-1);format time year4.; y=LOG(cons); dif=dif(y); cards; 262.7 328.8 356.1 364.0 424.0 441.6 481.2 556.5 595.4 537.7 543.7 544.8 572.7 590.1 632.8 679.1 649.2 698.2 728.8 776.9 853.5 917.7 967.4 1046.4 1099.0 1174.3 1264.9 1476.0 1794.0 2002.5 2181.5 2426.1 2899.2 3801.4 4374.0 5115.0 6534.6 7074.2 7250.3 8245.7 9704.8 12462.1 16264.7 ; proc gplot;plot cons*time=1 y*time=2 dif*time=3; symbol1 v=circle i=join c=black; symbol2 v=dot i=join c=red; symbol3 v=star i=join c=green; proc arima data=b;identify var=dif nlag=22;estimate p=1 method=ml; run;

模型为取对数以后的IAR(1)

ln( xt ) 0.105732 ( 1 0.53006B ) t

第3题非参数统计 data ex; do a=1 to 2; do i=1 to 11;input x @@; output;end;end; cards; 10.2 9.6 9.2 10.6 10.7 10.6 9.5 9.8 8.8 10.1 10.2 9.8 ; proc npar1way wilcoxon; class a; var x; run;

9.910.3

10.29.3

10.610.5

10.010.0

11.210.6

自己读结果，结果表明无差异

非参数检验第4题

data ex;do a=1 to 4;input n@@; do i=1 to n;input x@@; output;end;end; cards; 7 60 75 62 76 73 98 86 7 72 52 68 82 74 64 87 8 61 85 78 66 70 59 69 79 8 63 58 65 71 84 77 80 89 ; proc npar1way wilcoxon; class a;var x;run;

由上述结果可知，无显著差异

第5题

data ex; input x y @@; cards; 9 7.8 16.6 19.3 16.2 20.1 11.3 7.1 16.2 13 7.1 4.8 7.8 8.9 4 7.4 11.2 10 1.3 1.5 ; proc corr pearson spearman kendall; run;

由结果可知， 三个相关系数 分别为 0.88,0.82,0.67