如何对多项式回归曲线做confidence band

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

我的CODE如下

proc reg data=table2;
model daysofTm90=centeryear year3  /clb;
output out=model_set p=yfit;
run ;
title 'regression analysis for degree=3 ';
proc gplot data=model_set;
plot daysofTm90*year yfit*year/overlay;
symbol1 v=star c=blue I=none;
symbol2 v=none c=red I=spline;
run;

我要在画出的曲线上加上95%置信区间阴影，CODE该怎么改啊。谢谢

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：confidence dence band con BAN 多项式 如何

proc sgplot data=sashelp.classfit;
  title "Fit and Confidence Band from Precomputed Data";
  band x=height lower=lower upper=upper /
       legendlabel="95% CLI" name="band1";

hamsik11 发表于 2013-6-18 14:55
proc sgplot data=sashelp.classfit;
  title "Fit and Confidence Band from Precomputed Data";
  band ...

1117     proc tpspline data=model_set;
1118        model daysofTm90=centeryear year3/alpha = 0.05;
1119        output out =result pred uclm lclm;
1120     run;

ERROR: At least one smoothing variable is required. If you only want to fit a parametric model,
       you can use PROC GLM or PROC REG.
NOTE: The SAS System stopped processing this step because of errors.

WARNING: The data set WORK.RESULT may be incomplete.  When this step was stopped there were 0
         observations and 0 variables.
WARNING: Data set WORK.RESULT was not replaced because this step was stopped.

NOTE: PROCEDURE TPSPLINE used (Total process time):
      real time           0.00 seconds
      cpu time            0.00 seconds

1121
1122
1123   ods graphics on;
1124


1125     proc sgplot data=result;
1126        title "polynomial regression for Tm90";
1127
1128        xaxis label="year";
1129        yaxis label="days of Tm90";
1130
1131        band x=year lower=lclm_daysofTm90 upper=uclm_daysofTm90/name="range"
ERROR: Variable LCLM_DAYSOFTM90 not found.
ERROR: Variable UCLM_DAYSOFTM90 not found.
1132                    legendlabel="95% CI of Predicted daysofTm90";
1133        scatter x=year y=daysofTm90/name="obs" legendlabel="daysofTm90";
ERROR: Variable DAYSOFTM90 not found.
1134        series x=year y=p_daysofTm90/name="pred"
ERROR: Variable P_DAYSOFTM90 not found.
1135                    legendlabel="predicted values of daysofTm90"
1136                    lineattrs=graphfit(thickness=1px);
1137
1138        discretelegend "range" "obs" "pred";
1139     run;

我用了这些,log 出现错误怎么回事情啊，求高人指点