Question

我尝试通过proc univariate创建直方图。我们的目标是使用0到1.5范围内的0.1宽度的条板箱分配分布，然后将所有剩余的条板箱放入一个条板箱。

我应用了以下代码来标识从0到1.5的范围，而它无法管理其余部分。我该如何纠正代码？

proc univariate data=HAVE;
where pred between 0 and 1.5;
var pred;
histogram pred/ vscale=percent midpoints=0 to 2 by 0.1 normal (noprint);
run;

Answer 1

您可以尝试通过类似的以下代码来通过从一个变量创建两个变量来组合两个直方图：

/*Temporary DS with values ranging from 01. to 2.0*/
data have;
do i=0.1 to 2.0 by 0.1;
    output;
end;
rename i=pred;
run;

/*Creating two variables x(0.1-1.5) and y(1.6-2.0)*/
data have;
 set have;
 if pred<1.6 then x=pred;
 else y=pred;
 drop pred;
run;

/*Combine two Histograms*/
proc sgplot data=have;
   histogram x / nbins=15 binwidth=0.1;
   density x / type=normal;
   histogram y / nbins=5 binwidth=1.0;
   density y / type=normal;
   keylegend / location=inside position=topright noborder across=2;
   xaxis display=(nolabel) values=(0.1 to 2.5 by 0.1);
run;

Answer 2

创建自己的网上论坛
创建一种格式，以显示您想要的方式

用SGPLOT绘制

*create your own groups for data, especially the last group;
data mileage;
set sashelp.cars;
mpg_group=floor(mpg_highway / 10);

if mpg_group in (5, 6, 7) then
    mpg_group=5;
keep mpg_highway mpg_group;
run;

*format to control display;

proc format;
value mpg_fmt 1='0 to 10' 2='11 to 20' 3='21 to 30' 4='31 to 40' 5='40+';
run;

*plot the data;
proc sgplot data=mileage;
vbar mpg_group /stat=freq barwidth=1;
format mpg_group mpg_fmt.;
run;

SAS ---如何创建在特定范围内具有特定bin的直方图图

2 个答案: