Use the UNIVAR macro.

1. Issue the following commands to SAS (with appropriate
   substitutions for "U:\").

LIBNAME GFLib 'U:\dm\sasdata' ; 

LIBNAME RCLib 'U:\dm\RC' ;

2. Read your Excel file into SAS with the EXCELSAS macro. 
   	Note that this step only needs to be done once.  
   In other words, if you close SAS down today but decide 
   that you want to apply the UNIVAR macro to the same 
   data set on another day, you do not need to read your 
   Excel file into SAS a second time.  This is because the 
   EXCELSAS macro actually creates a permanent SAS dataset.

3. Apply the UNIVAR macro.  Detailed instructions on how  
   to fill in the blanks are given on pages 60-63 of the 
   textbook; I will make a few comments below.  I used the 
   following for the illustration in Lecture 3.
	RCLib.saheart		sbp 
	chd			95	
	subject			1	
	U:\dm\RC\		U:\dm\RC\
	word
   Comments:  You can put any continuous variable from the
   data set in the second field.  This variable does not have  
   to be what you would consider the response variable.  You 
   can also put more than one variable in the second field, 
   if you separate them by spaces.  Since the length of the
   second field is limited, you may need more than one
   invocation of the UNIVAR macro if you are interested in
   looking at several continuous variables.
	You can put any categorical variable from the data
   set in the third field, or you can leave the third field
   blank.  The former option performs exploratory data 
   analysis on the continuous variable(s) within strata 
   determined by the categorical variable, while the latter
   option does not entail any stratification.  
	What I have above looks at the distribution of 
   systolic blood pressure among subjects with coronary 
   heart disease and at the distribution of systolic blood 
   pressure among subjects without coronary heart disease.

4. Based on the exploratory data analysis, you can define
   transformed versions of the continuous variable(s) and
   remove extreme outliers that are mistakes for which
   corrections are not possible.  The macro creates new data 
   sets with outliers removed, but I would not use these since 
   the macro discards moderate outliers (which are often valid
   observations) along with extreme outliers (which may be
   mistakes).  Here is sample code that creates a new SAS
   data set {saheartclean.sas7bdat} starting from 
   {saheart.sas7bdat} by removing all observations for 
   which  sbp > 200  and defining new variables as the 
   logarithm and square root of systolic blood pressure.  

	DATA RCLib.saheartclean;
	SET RCLib.saheart;
	if (sbp > 200) then delete;
        logsbp = log(sbp);
        sqrtsbp = sqrt(sbp);
	RUN;

5. You may want to apply the UNIVAR macro again to see how
   the transformed versions of the continuous variable(s)
   turn out.
	RCLib.saheartclean	logsbp 
	chd			95	
	subject			2	
	U:\dm\RC\		U:\dm\RC\
	word
 

Use the FREQ (or FREQUENCY) macro.

1. Issue the following commands to SAS (with appropriate
   substitutions for "U:\").

LIBNAME GFLib 'U:\dm\sasdata' ; 

LIBNAME RCLib 'U:\dm\RC' ;

2. Read your Excel file into SAS with the EXCELSAS macro. 

3. Apply the FREQ (or FREQUENCY) macro.  Detailed instructions 
   on how to fill in the blanks are given on pages 53-55 of 
   the textbook; I will make a few comments below.  I used the 
   following for the illustration in Lecture 3.
	RCLib.saheart		chd
	famhist
	freq	 		midpoint
	color			3
	U:\dm\RC\		U:\dm\RC\
	word
   Comments:  Use the second field for the categorical variable
   whose distribution you want to examine.  Optionally, use the
   third and fourth fields for other categorical variables by
   which you want to stratify.
	What I have above looks at the distribution of coronary
   heart disease among subjects with a family history and among
   subjects without a family history.