Data sets and R code

SPSS v11 data sets in zipped format, can be imported in R and other programs. Use of the data sets is strictly for educational purposes. If research is considered, please contact me or the primary researchers.

• All.sav: files in one zip file
• Finlayson example: Fmort.sav with mortality in over 1,000,000 surgical procedures, use a weight variable
• Parts of the GUSTO-I data set: West region (n=2188), sample4 (n=785), sample5 (n=429), and sample2 (n=259, March 2014)
• Aneurysm example: aneurysm.sav (n=238)
• Testicular cancer example: t821.sav (n=1094, 821 with LDH measured)
• Outcome of traumatic brain injury: TBI.sav (n=2159)
• Survival data for ch 23: SMART.sav (n=3873)
  

Part I: Prediction models

• 4. Competing risks models (see also chapter04)
• 5. Overfitting and optimism
• 6. Choosing between statistical models
  

Part II: Developing valid prediction models

• 7. Missing values
• 8. Case study on missing values
• 12. Assumptions of regression models
• 13. Shrinkage
• 14. Estimation with external information
• 15 and 16. Evaluation of Performance
• 17. Validation of prediction models: performance of the bootstrap
  

Part IV: Applications

• 23. Case study on survival analysis
  

Code with published papers

• Stat Med 2000 Comparison modeling strategies
• Stat Med 2004 Validation and updating
• Q J Med 2010 Comparing and ranking hospitals with R code
• Epidemiology 2010 Performance measures with R code
  
• EurJClinInv 2011 Evaluation of markers
  
• RevEspCard 2011 Evaluation of models and markers
  
• R code for EHJ 2014 tutorial: seven steps for development and an ABCD for validation with gusto data
  
  
• Windows and Mac modified versions of val.prob functions with
  
  1. Van Calster JCE 2016 Calibration hierarchy paper, which has a
  2. technical Appendix and a
  3. description of the modified val.prob.ci.2 function
     

The package can be installed in R as:
require(devtools)
install_git(“https://github.com/BavoDC/CalibrationCurves”)
It produces nice graphics:

Some basic background to R
learn R at UCLA
learn R at Amazon
R Packages, Functions, and Documentation by Frank Harrell