*Copyright @ www.mycsg.in;

Numeric functions to analyse variables in SAS

Function	Definition
N	Counts the number of non-missing numeric values in a list of variables.
NMISS	Counts the number of missing numeric values in a list of variables.
SUM	Returns the sum of non-missing numeric values in a list of variables.
MEAN	Returns the average of non-missing numeric values in a list of variables.
MIN	Returns the smallest non-missing numeric value in a list of variables.
MAX	Returns the largest non-missing numeric value in a list of variables.

Create sample data

We create a dataset with three score variables
Some values are intentionally missing, so the effect of functions such as `n` and `nmiss` can be seen clearly

data scores;
   infile datalines truncover;
   input score1 score2 score3;
datalines;
95 88 72
. 70 80
85 . .
. . .
50 60 70
;
run;

Inspect `scores` and note which rows contain complete values and which rows contain missing values

Example of the N function

The `n` function counts how many listed variables have non-missing numeric values on the current observation
This is useful when checking how much information is present on each row

data scores01;
    set scores;
    answered = n(score1, score2, score3);
run;

Review `answered` and confirm that rows with more available scores have larger counts

Example of the NMISS function

The `nmiss` function counts how many listed variables are missing on the current observation
This is the complement of the `n` count for the same variable list

data scores01;
    set scores;
    notanswered = nmiss(score1, score2, score3);
run;

Compare `notanswered` with `answered` and verify that their totals reflect the three score variables on each row

Example of the SUM function

The `sum` function adds the non-missing values in the list of variables
Missing values are ignored rather than causing the result to become missing
If all listed values are missing, the result is missing

data scores01;
    set scores;
    sumscores = sum(score1, score2, score3);
run;

Inspect `sumscores` and note how rows with partial missing data still get a total based on available values

Example of the MEAN function

The `mean` function calculates the average of non-missing values only
The denominator is the number of available values, not the total number of variables listed

data scores01;
    set scores;
    meanscore = mean(score1, score2, score3);
run;

Compare `meanscore` across rows and observe how missing values are excluded from the average calculation

Example of the MIN function

The `min` function returns the smallest non-missing value from the list
This is useful when finding the lowest score or earliest numeric measure across related variables

data scores01;
    set scores;
    minscore = min(score1, score2, score3);
run;

Inspect `minscore` and confirm that it reflects the lowest available score on each observation

Example of the MAX function

The `max` function returns the largest non-missing value from the list
This is useful when finding the highest score or greatest value across related variables

data scores01;
    set scores;
    maxscore = max(score1, score2, score3);
run;

Inspect `maxscore` and confirm that it reflects the highest available score on each observation

Compare several numeric functions together in one step

In practical programming, we often create multiple derived metrics from the same set of variables in one DATA step
This final example brings the functions together so the learner can compare them side by side

data scores_summary;
    set scores;
    answered    = n(score1, score2, score3);
    notanswered = nmiss(score1, score2, score3);
    sumscores   = sum(score1, score2, score3);
    meanscore   = mean(score1, score2, score3);
    minscore    = min(score1, score2, score3);
    maxscore    = max(score1, score2, score3);
run;

Review `scores_summary` and compare all derived variables on the same row
This makes the role of each numeric function easier to understand in relation to the others

Key points to remember

`n` and `nmiss` count available and missing numeric values
`sum` and `mean` ignore missing values instead of failing immediately
`min` and `max` return the smallest and largest non-missing values
These functions are especially useful when multiple related numeric variables must be analysed within the same observation

*Copyright @ www.mycsg.in;

Numeric functions to analyse variables in SAS

Create sample data

SAS Log

Dataset View

Example of the N function

SAS Log

Dataset View

Example of the NMISS function

SAS Log

Dataset View

Example of the SUM function

SAS Log

Dataset View

Example of the MEAN function

SAS Log

Dataset View

Example of the MIN function

SAS Log

Dataset View

Example of the MAX function

SAS Log

Dataset View

Compare several numeric functions together in one step

SAS Log

Dataset View

Key points to remember