ST500                               Third Session

_sas syntax + assignment                 {remember: man sas>f1}
  /*coment is listed here */
To execute: sas command &
             This creates command.log and command.lst  as default
             file names.
The '&' at the end of the command line means run the command line in
'background'.

_ph(phone book @ CSU) _netscapei or lynx: home-page for st500
   http://www.stat.colostate.edu/~zumbrunn/St500/


_spss commands:syntax + examples         {remember: man spss>f1}
   COMMENT 'comment is listed here'
To execute: spss -m command>command.out &
        spss -m -s1m command>command.out & (if spss says it needs
        more memory)


To read the data file for assignment 1 (b):

data list file='assign1.dat' list/
   name (a10) sex (a1) height weight age {NOTE: when you indent too far
missing values weight(-1)           with vi, it inserts 'tabs' that
comment note the missing value statement above     spss does NOT like
on your list cases                          continuation lines where
it is frequencies variables=name to age   looking for a blank!} sort
cases by sex split file by sex descriptive name to age/statistics save
outfile=sclass select if sex='m' save outfile=smale get file=sclass select
if sex='f' save outfile=sfemale add files file=smale/file=sfemale ...
...  compute Iwt=lg10(weight) examine variables= height to Iwt /
   plot=stemleaf npplot boxplot
...   ...





ST500     Assignment 1 (b) - spss

To execute SPSS with a control file: spss -m control>control.out& NOTE:
Since we are running spss in 'batch' mode, there are no 'periods' required
at the end of each statement. IF we were to run interactively a 'period'
is required to end each statement!  (e.g. SAVE OUTFILE=SMALE would need
to be SAVE OUTFILE=SMALE.)  You may check the ../assign/a1b file to get
additional 'help'!

The following tasks should be done with SPSS using the assign1.dat
data file. You need email to me (st500in1) the (1)control statements
(2) control.out file before the beginning of class   on Jan 8

1. Read assign1.dat into SPSS then list it. (DATA LIST FILE='..'/.. \
LIST CASES \ EXECUTE) 2. Do a frequency distribution of all data
fields. (FREQUENCIES ..) The negative value for weight should be 'missing'
(MISSING VALUES ..)  3. Compute means/minimums/maximums of height,weight
and age for sex (m and f) separately.  (SORT CASES BY SEX \ SPLIT FILE
BY SEX \ DESCRIPTIVE ..)  4. Create two SPSS data files: one with m and
one with f. (SAVE OUTFILE=SCLASS \ SELECT IF SEX='m' \ SAVE OUTFILE=SMALE
\ GETFILE=SCLASS \ SELECT IF ..\ SAVE OUTFILE=..)  5. Recombine the two
files into one file and print.  (ADD FILES FILE=SMALE/FILE=SFEMALE \ LIST
CASES \ EXECUTE) 6. Sort the file by name within sex, and print. (SORT
CASES BY .... \ LIST CASES \ EXECUTE) 7. Plot weight against height and
age, using two symbols on the plot to indicate sex. (PLOT PLOT=WEIGHT WITH
.... BY ..  8. compute LOG10 of weight (Iwt), then plot Iwt against age
with symbols for sex as in 7.  (COMPUTE Iwt=lg10(weight) \ PLOT PLOT=..)
9. Use EXAMINE to get stem + Leaf plot, boxplot and normal probability
plot of weight, height, and Iwt.

                                  (EXAMINE VARIABLES= HEIGHT TO LWT
                                  /PLOT=STEMLEAF NPPLOT BOXPLOT)

                        10. SPSS can generate data, but it's a little
                        'awkward'.
                             INPUT PROGRAM
LOOP #I=1 to 200 COMPUTE A=NORMAL(1) COMPUTE B=A*.2+1 COMPUTE C=B*B
COMPUTE D=10**B END CASE END LOOP END FILE END INPUT PROGRAM EXAMINE
VARIABLES=A TO D/PLOT=STEMLEAF NPPLOT BOXPLOT