一、单变量数据探索
1、均值、中位数、众数、方差、标准差、标准误差
![59c9a873e40ea6f9df181d3e668ef2d1.png](https://img-blog.csdnimg.cn/img_convert/59c9a873e40ea6f9df181d3e668ef2d1.png)
对连续性变量进行数据探索主要用means和univariate程序;对离散性变量进行数据探索主要用freq。
![11f294fe42b6518173cec2d34ca1baaa.png](https://img-blog.csdnimg.cn/img_convert/11f294fe42b6518173cec2d34ca1baaa.png)
proc means data=data.b_rise maxdec=4;
var weight;
title 'Descriptive Statistics for WEIGHT';
run;
proc means data=data.b_rise
maxdec=4
n mean median std var q1 q3;
var weight;
title 'Selected Descriptive Statistics for weight';
run;
proc univariate data=data.b_cereal;
class brand;
var weight;
probplot weight / normal
(mu=est sigma=est color=blue w=1);
title 'Univariate Analysis of the Cereal Data';
run;
proc sort data=data.b_cereal out=b_cereal;
by brand;
run;