FORMAT statements can be used to change the look of your output

1

• FORMAT statements can be used to change the look
  of your output
• if FORMAT is in the DATA step, then the formats
  are permanent and stored with the dataset
• if FORMAT is used in a PROC step, then the
  formats are temporary and are not in play after
  that PROC executes
• PUT statements are used to write data to a file
  or to write reports formats can be used (along
  with other pointer controls) within the PUT
  try this with the data on page 109

2

• DATA sales
• INFILE 'c\MyRawData\Candy.dat'
• INPUT Name 1-11 Class _at_15 DateReturned
  MMDDYY10. CandyType Quantity
• Profit Quantity 1.25
• PROC SORT DATA sales
• BY Class
• PROC PRINT DATA sales
• BY Class
• SUM Profit
• VAR Name DateReturned CandyType Profit
• FORMAT DateReturned DATE9. Profit DOLLAR6.2
• TITLE 'Candy Sales for Field Trip by Class'
• FOOTNOTE USING FORMATS
• RUN

3

• DATA FOR THE PREVIOUS SLIDE
• Adriana 21 3/21/2000 MP 7
• Nathan 14 3/21/2000 CD 19
• Matthew 14 3/21/2000 CD 14
• Claire 14 3/22/2000 CD 11
• Caitlin 21 3/24/2000 CD 9
• Ian 21 3/24/2000 MP 18
• Chris 14 3/25/2000 CD 6
• Anthony 21 3/25/2000 MP 13
• Stephen 14 3/25/2000 CD 10
• Erika 21 3/25/2000 MP 17
• USE THIS DATA TO WRITE THE NAME, QUANTITY SOLD
  AND PROFIT (IN DOLLAR FORMAT) TO A TEXT FILE
  NAMED CANDYOUT.TXT - try this.
• PUT NAME _at_15 QUANTITY _at_18 PROFIT DOLLAR8.2

4

• This doesnt do the job the PUT statement
  without a FILE statement writes in the LOG
  window
• So add the following in the DATA step (can go
  after the INFILE for example)
• FILE path to file goes here
• Check out section 4.6 on pages 110-111 for a
  small list of the standard formats that SAS
  uses
• If the FORMAT you want is not available you may
  write your own with PROC FORMAT

5

• PROC FORMAT creates your own formats that can
  later be associated with output through the use
  of a FORMAT statement (dont forget the end dot
  when they are actually used)
• The syntax requires one or more VALUE statements
  after the beginning PROC FORMAT statement. The
  VALUE statements give names for the formats (name
  must begin with if format is for character
  data) and tell what values of the variable will
  be associated with the formats
• PROC FORMAT VALUE name range format
  value
• Character values being formatted must be in
  quotes. Ranges of values can be indicated in
  several ways (see bottom of p. 112)

6
PROC FORMAT VALUE island siShell
Island ph Pizza Hut VALUE mass 0 - lt
5 small 5-high large NOTE For each
variable in SAS there is a construct the
variable measures a name you give to the
variable the values of the variable You may
change the name for outputting with LABEL You may
change the values for outputting with FORMAT
(either SAS built-in ones or ones you write
yourself with PROC FORMAT) See the example from
p. 113
7
DATA carsurvey INPUT Age Sex Income Color
datalines 19 1 14000 Y 45 1 65000 G 72 2 35000
B 31 1 44000 Y 58 2 83000 W PROC FORMAT VALUE
gender 1 'Male' 2 'Female'
VALUE agegroup 13 -lt 20 'Teen'
20 -lt 65 'Adult' 65 - HIGH
'Senior' VALUE col 'W' 'Moon White'
'B' 'Sky Blue' 'Y'
'Sunburst Yellow' 'G' 'Rain
Cloud Gray' PROC PRINT DATA carsurvey
FORMAT Sex gender. Age agegroup. Color col.
Income DOLLAR8. run quit
8

• PROC MEANS options
• Other possible statements with PROC MEANS are
• BY varible-list (makes MEANS do separate
  analyses for each level of vars. in the list
  dataset must be SORTed by the variables in the
  list)
• CLASS variable-list (makes MEANS do as with BY,
  but has more compact output dataset does not
  have to be SORTed)
• VAR variable-list (makes MEANS only operate on
  the variables in the list if VAR statement is
  not there, then all numeric variables are
  analyzed)
• Show the difference between BY and CLASS

9

• PROC MEANS may be used to create new datasets
  with the OUTPUT statement these new datasets
  contain the resulting statistics (mean, sum,
  etc.) by doing the PROC MEANS. The following
  example creates a new dataset (named STATS)
  containing the means and standard deviations of
  the mass and height variables from both marshes
• PROC MEANS DATApadgett BY MARSH
• OUTPUT OUTstats
• MEAN(totmass plantht)massmean htmean
• STD(totmass plantht)masssd htsd run
• Go over the example on page 119 for another look
  at this important method of creating new datasets
  in SAS

10

• PROC FREQ does frequency distributions for single
  variables and cross tabulations for two or more
  variables together
• TABLES varible-combinations is a required
  statement
• the between two variables creates a variable
  combination and signals that a cross tabulation
  is to be done.
• options may be added to PROC FREQ by listing them
  after a slash (/) in the TABLES statement. Some
  key options for PROC FREQ are
• LIST which prints cross tabulations in list
  format instead of tabular form
• MISSING includes missing values in the
  distributions (and hence in the computations)
• NCOL and NROW supresses the printing of the
  column and row percentages respectively in cross
  tabulations
• OUTdataset_name creates a dataset of
  frequencies.
• TRY THESE OUT ON PADGETT DATA

11

• For Wednesday
• Read Chapter 4 up through 4.7, 4.9-4.11
• HW 1. write a series of PROC FORMATs so we may
  better understand the oscars dataset (e.g.,
  gender, born in u.s., etc) test them out and
  then send me the file with your SAS code 2. try
  out the new results for PROCs MEANS and FREQ on
  the same dataset
• Be prepared to work on reporting
• using the PUT statement along with FILE (4.8)
• using PROC TABULATE (4.12-4.16)
• using PROC REPORT (4.17-4.21)
• Start looking for your own dataset to analyze for
  the midterm