Skip to main content

Advertisement

Table 4 Complete case bootstrap model selection for the outcome measure persistent disability

From: The search for stable prognostic models in multiple imputed data sets

  most frequently selected models rank
Predictors* 1 2 2 4 5 B CCA
inability to perform daily activities X X X X X 1 1
both shoulders afflicted X - - X X 2 3
shoulder complaints in the past year X X X X - 3 2
concomitant lower extremity pain X X X X X 4 5
more disability at baseline X X X X X 5 6
concomitant lower back pain - - X X X - 4
older age - - - - - - -
longer duration of complaints - - - - - - -
acute onset - - - - - - -
Count 33 23 23 22 16   
% 6.6 4.6 4.6 4.4 3.2   
  1. *            - only those predictors that appeared in ≥40% of the first bootstrap model selection step are presented
  2. rank       - the order of appearance of predictors in the derived models arranged by their predictive ability (regression coefficient estimates)
  3. B           - the complete case date based bootstrap selected model (i.e. the most frequently occurring combination of predictors in 500 replicate data sets of the second bootstrap model selection step)
  4. CCA       - the complete case data based model derived without additional bootstrap was the fourth most occurring combination of predictors in the bootstrap model selection procedure
  5. Count     - the number of times the model was selected in the 500 replicate data sets of the second bootstrap model selection step