SUDAAN 9 Examples (Windows)
Below are examples from each of the SUDAAN procedures. Each example is packaged as a downloadable self-extracting executable. Each executable contains both standalone and SAS-callable versions of the program, data and output for the procedure. By default these files will be unzipped to the directory C:\SUDAAN Examples. All examples are already set up to work from this directory.
Data sets used in these examples are documented fully in the SUDAAN Example Manual in Section 1.8.
ALL Sample Examples
AllExamples.zip
Select AllExamples.exe to download the entire set of sample examples below in one step.
CROSSTAB
CROSSTAB.zip
This example uses PROC CROSSTAB to illustrate the estimation of odds ratios and performing stratified tests of association (CMH tests). The data is taken from the BRFSS study.
DESCRIPT
DESCRIPT.zip
This example uses PROC DESCRIPT to explore racial differences in the proportion of women who have had PAP smears and mammograms. The data is taken from the BRFSS study.
KAPMEIER
KAPMEIER.zip
This example illustrates the new KAPMEIER procedure. The EXERCISE data from Crouchley and Pickles is used.
LOGISTIC
LOGISTIC.zip
This example uses PROC LOGISTIC and the BRFSS data to illustrate the new CLASS statement, and the Hosmer-Lemeshow test for goodness of fit.
LOGLINK
LOGLINK.zip
This example uses PROC LOGLINK with the Thall and Vail data to illustrate GEE modeling, the EFFECTS statement, the CLASS statement and confidence limits for the parameter estimates.
MULTILOG
MULTILOG.zip
This example uses PROC MULTILOG with the NHANES III multiply imputed data sets to illustrate analysis of multiply imputed data using the MI_COUNT option and using the MI_VAR statement.
RATIO
RATIO.zip
In this example proc RATIO is used to estimate the average number of days per hospital episode incidence, with indirect standardization to a sex-by-race distribution. A subset of the 1985 HIS data set is used for this analysis.
REGRESS
REGRESS.zip
Here we illustrate the use of GEE with independent vs. exchangeable working correlations in REGRESS, for robust variance estimation. We also demonstrate the CLASS, REFLEVEL and CONDMARG statements. The Butler (Teratology) Data is used in this analysis.
SURVIVAL
SURVIVAL1.zip
This example illustrates the SURVIVAL procedure with the discrete proportional hazards model using NHANES I and Longitudinal Follow-Up Data.
SURVIVAL2.zip
This example illustrates the Cox proportional hazards model in SURVIVAL using this same data set. This example uses Efron’s tie-handling, and illustrates use of both WR and Delete-Jackknife designs.
SURVIVAL3.zip
This example uses the exercise data from Crouchley and Pickles, and illustrates the PREDMARG statement with continuous variables, and the new Martingale residuals.