The UNIVARIATE Procedure

The OUTTABLE= data set saves univariate statistics in a data set that contains one observation per analysis variable. The variables shown in Table 39 are saved.

Table 39: Variables in the OUTTABLE= Data Set

Variable	Description
_CSS_	Corrected sum of squares
_CV_	Coefficient of variation
_GEOMEAN_	Geometric mean
_GINI_	Gini’s mean difference
_HARMEAN_	Harmonic mean
_KURT_	Kurtosis
_MAD_	Median absolute difference about the median
_MAX_	Maximum
_MEAN_	Mean
_MEDIAN_	Median
_MIN_	Minimum
_MODE_	Mode
_MSIGN_	Sign statistic
_NMISS_	Number of missing observations
_NOBS_	Number of nonmissing observations
_NORMAL_	Test statistic for normality
_P1_	1st percentile
_P5_	5th percentile
_P10_	10th percentile
_P90_	90th percentile
_P95_	95th percentile
_P99_	99th percentile
_PROBM_	p-value of sign statistic
_PROBN_	p-value of test for normality
_PROBS_	p-value of signed rank test
_PROBT_	p-value of t statistic
_Q1_	25th percentile (lower quartile)
_Q3_	75th percentile (upper quartile)
_QN_
_QRANGE_	Interquartile range (upper quartile minus lower quartile)
_RANGE_	Range
_SGNRNK_	Centered sign rank
_SKEW_	Skewness
_SN_	(see Robust Estimates of Scale)
_STD_	Standard deviation
_STDGINI_	Gini’s standard deviation
_STDMAD_	MAD standard deviation
_STDMEAN_	Standard error of the mean
_STDQN_	standard deviation
_STDQRANGE_	Interquartile range standard deviation
_STDSN_	standard deviation
_SUMWGT_	Sum of the weights
_SUM_	Sum
_T_	Student’s t statistic
_USS_	Uncorrected sum of squares
_VARI_	Variance
_VAR_	Variable name

The OUTTABLE= data set and the OUT= data set (see the section OUT= Output Data Set in the OUTPUT Statement) contain essentially the same information. However, the structure of the OUTTABLE= data set might be more appropriate when you are computing summary statistics for more than one analysis variable in the same invocation of the UNIVARIATE procedure. Each observation in the OUTTABLE= data set corresponds to a different analysis variable, and the variables in the data set correspond to summary statistics and indices.

For example, suppose you have 10 analysis variables (P1-P10). The following statements create an OUTTABLE= data set named Table, which contains summary statistics for each of these variables:

data Analysis;
   input A1-A10;
   datalines;
 72  223  332  138  110  145   23  293  353  458
 97   54   61  196  275  171  117   72   81  141
 56  170  140  400  371   72   60   20  484  138
124    6  332  493  214   43  125   55  372   30
152  236  222   76  187  126  192  334  109  546
  5  260  194  277  176   96  109  184  240  261
161  253  153  300   37  156  282  293  451  299
128  121  254  297  363  132  209  257  429  295
116  152  331   27  442  103   80  393  383   94
 43  178  278  159   25  180  253  333   51  225
 34  128  182  415  524  112   13  186  145  131
142  236  234  255  211   80  281  135  179   11
108  215  335   66  254  196  190  363  226  379
 62  232  219  474   31  139   15   56  429  298
177  218  275  171  457  146  163   18  155  129
  0  235   83  239  398   99  226  389  498   18
147  199  324  258  504    2  218  295  422  287
 39  161  156  198  214   58  238   19  231  548
120   42  372  420  232  112  157   79  197  166
178   83  238  492  463   68   46  386   45   81
161  267  372  296  501   96   11  288  330   74
 14    2   52   81  169   63  194  161  173   54
 22  181   92  272  417   94  188  180  367  342
 55  248  214  422  133  193  144  318  271  479
 56   83  169   30  379    5  296  320  396  597
;

proc univariate data=Analysis outtable=Table noprint;
  var A1-A10;
run;

The following statements create the table shown in Figure 15, which contains the mean, standard deviation, and so on, for each analysis variable:

proc print data=Table label noobs;
   var _VAR_ _MIN_ _MEAN_ _MAX_ _STD_;
   label _VAR_='Analysis';
run;