UCLA Academic Technology Services HomeServicesClassesContactJobs
Stat Computing > Statistica > Textbook Examples > Computer-Aided Multivariate Analysis, 4th
 Search

Statistica Textbook Examples
Computer-Aided Multivariate Analysis by Afifi, Clark and May
Chapter 13: Regression analysis using survival data

We have skipped pages 336-342 for now.

Page 347 Table 13.3  Percentage of deaths versus explanatory variables

NOTE:  For each table, change the variable in List 1 to the variable for that table.

File
  Open - open the surv data set
Statistics
  Basis statistics
    Tables and banners
      OK
        "Crosstabulation" tab
          Specify tables - select staget for List 1 and death for list 2
            OK
              OK
                "Options" tab - uncheck "Highlight counts" and check "Percentages of row counts"
                  Summary
Summary Frequency Table (surv.sta)
Table: STAGET(2) x DEATH(2)
  STAGET DEATH alive DEATH dead Row Totals
Count small 122 91 213
Row Percent   57.3% 42.7%  
Count large 75 113 188
Row Percent   39.9% 60.1%  
Count All Grps 197 204 401


Summary Frequency Table (surv.sta)
Table: PERFBL(2) x DEATH(2)
  PERFBL DEATH alive DEATH dead Row Totals
Count good 174 163 337
Row Percent   51.6% 48.4%  
Count poor 22 40 62
Row Percent   35.5% 64.5%  
Count All Grps 196 203 399


Summary Frequency Table (surv.sta)
Table: TREAT(2) x DEATH(2)
  TREAT DEATH alive DEATH dead Row Totals
Count saline 99 96 195
Row Percent   50.8% 49.2%  
Count BCG 98 108 206
Row Percent   47.6% 52.4%  
Count All Grps 197 204 401

Summary Frequency Table (surv.sta)
Table: POINF(2) x DEATH(2)
  POINF DEATH alive DEATH dead Row Totals
Count no 191 189 380
Row Percent   50.3% 49.7%  
Count yes 5 15 20
Row Percent   25.0% 75.0%  
Count All Grps 196 204 400

Page 348 Table 13.4  Log-linear model for lung cancer data:  results

NOTE:  We were unable to reproduce this table.

Page 349 Table 13.5  Cox's model for lung cancer data:  results

Statistics
  Advanced linear/nonlinear models
    Survival analysis
      Regression models
        OK
          "Quick" tab
            Variables - select days for survival, staget, perfbl, poinf and treat 
            as independent variables and death as the censoring variable
              OK
                Code for complete responses - double click on the field and select "dead"
                  Code for censored responses - double click on the field and select "alive"
                    OK
                      Summary
Dependent Variable: DAYS (surv.sta) Censoring var.: DEATH
Chi-square = 28.5333 df = 4 p = .00001
  Beta Standard Error t-value exponent beta Wald Statist. p
STAGET 0.54 0.14 3.82 1.72 14.6 0.00
TREAT 0.08 0.14 0.55 1.08 0.3 0.58
PERFBL 0.53 0.19 2.88 1.71 8.3 0.00
POINF 0.67 0.28 2.38 1.95 5.7 0.02

We have skipped pages 350-354 for now.

Page 357 Figure 13.9  Computer-generated graph of log(-logS(t)) versus t for lung cancer data (A = large tumor, B = small tumor)

NOTE:  We were unable to reproduce this graph.

How to cite this page

Report an error on this page

UCLA Researchers are invited to our Statistical Consulting Services
We recommend others to our list of Other Resources for Statistical Computing Help
These pages are Copyrighted (c) by UCLA Academic Technology Services

The content of this web site should not be construed as an endorsement of any particular web site, book, or software product by the University of California