Exercises

Topics for exercises are listed here; programming code and answers are available from the author

• R code used for validation of models from small parts of GUSTO-I (practical Utrecht 2011, Gusto data)
  
• Simulate optimism for 20 random variable with 50 events
• Study overfitting and optimism (see chapter05)
• Study the relevance of linearity and interactions in prediction modeling (see chapter06)
• Study the prognostic value of predictors of outcome in TBI with background document and data set
• Validate prognostic models in TBI (from US to International) with background document and data set
  
• Perform single and multiple imputation in TBI data set
• Study non-linearity in testis cancer example
• Study performance of testis cancer prediction models
• Develop a valid model with GUSTO-I data (see chapter22)
• Perform updating with GUSTO-I data