Quick Reference Sheet
I. Data set in SAS Data Step
A. Data set must:
1. Have the following ordering of its variables:
Time, Censoring Indicator, Covariates.
2. Be sorted by ascending time.
B. The censoring indicator needs to specified as follows:
0 - censored; 1 - event.
C. All categorical variables should be defined as binary variables.
D. Only fixed covariates should be used (not time-dependent).
II. Proc IML
A. User now needs to be in IML, which is done by issuing the statement "proc iml" in the program.
B. Additive macro is included in the program via the "%include 'filename'" statement where 'filename'
is the name of the additive macro file.
C. Eight parameters are defined
1. Data set parameter: name of user's data set.
2. Confidence level parameter: defines alpha for (1-alpha)*100% confidence intervals.
3. Time unit parameter: defines the units of the time variable.
4. Variable list parameter: lists the names of the covariates, in the order they are read in
the data set.
5. Option parameter: defines which options of the five possible options the user wants by
placing a "y" for "yes" or a "n" for "no" in the position of the desired option.
Options:
(1) Printing of the parameter estimates and respective standard deviations for each
event time.
(2) Testing of general linear contrasts.
(3) Output data set containing parameter estimates, standard deviations, and
confidence limits.
(4) Line plots of each parameter estimate versus time.
(5) Output data set containing U and V for test statistics.
6. Contrast matrix parameter: r*(p+1)-matrix that defines the r linear contrasts to be tested.
7. Parameter naming the first output data set: contains the output data set requested from
Option (3) above.
8. Parameter naming the second output data set: contains the output data set requested from
Option (5) above.
D. Macro invocation using the following statement:
%additive(data set parameter,confidence level parameter,time unit parameter, variable list
parameter, option parameter, contrast matrix parameter, parameter naming the first output
data set,parameter naming the second output data set).
E. Contents of first output data set
1. Contains parameter estimates over time as well as the respective standard deviations and
(1-alpha)*100% confidence intervals.
a. First column is time.
b. Next (p+1) columns are the parameter estimates (in order).
c. Next (p+1) columns are the respective standard deviations.
d. Next (p+1) columns are the respective lower confidence limits.
e. Final (p+1) columns are the respective upper confidence limits.
F. Contents of second output data set
1. Contains the vector U and the respective variance-covariance matrix V.
a. First column is the individual Uj.
b. Next p columns are the variance-covariance matrix V.
56