【SAS NOTES】proc univariate检查单变量分布

1 proc univariate data=mysas.mmstwo;
2     var wangnei;
3 run;

结果

The SAS System 17:32 Saturday, February 14, 2009 1

The UNIVARIATE Procedure
Variable: wangnei

Moments

N 48 Sum Weights 48
Mean 8135197.19 Sum Observations 390489465
Std Deviation 1872000.61 Variance 3.50439E12
Skewness 1.15120577 Kurtosis 0.71878439
Uncorrected SS 3.34141E15 Corrected SS 1.64706E14
Coeff Variation 23.0111277 Std Error Mean 270200.014

Basic Statistical Measures

Location Variability

Mean 8135197 Std Deviation 1872001
Median 7713962 Variance 3.50439E12
Mode . Range 7438151
Interquartile Range 2020383

Tests for Location: Mu0=0

Test -Statistic- -----p Value------

Student's t t 30.10806 Pr > |t| <.0001
Sign M 24 Pr >= |M| <.0001
Signed Rank S 588 Pr >= |S| <.0001

Quantiles (Definition 5)

Quantile Estimate

100% Max 13050028
99% 13050028
95% 12280391
90% 11480186
75% Q3 8849826
50% Median 7713962
25% Q1 6829443
10% 6234089
5% 6185196
1% 5611877
0% Min 5611877

Extreme Observations

------Lowest----- ------Highest-----

Value Obs Value Obs

5611877 11 11480186 40
5728596 9 11605647 41
6185196 13 12280391 34
6186945 16 12970193 37
6234089 12 13050028 46

skewness 偏度。

kurtosis 峰度

正态分布的skewness and kurtosis is 0

2、可以为该变量分布创建图形，并标注不同的分布标准

 1 proc univariate data=mysas.mmstwo;
 2     var wangnei;
 3 run;
 4 proc univariate data=mysas.mmstwo;
 5     var wangnei;
 6     probplot wangnei;
 7 run;
 8 proc univariate data=mysas.mmstwo;
 9     var wangnei;
10     histogram wangnei/normal;
11 run;

可创建的图形有cdfplot histogram ppplot probplot qqplpt

对比的标准分布有beta exponential gamma lognormal normal weibull

指定图像输出目录：

1 ods graphics on;
2 ods listing gpath='C:\Documents and Settings\My Documents\My SAS Files';