Skip to main content

Advertisement

Table 2 Some ideas for the improvement of benchmarking practice

From: Towards evidence-based computational statistics: lessons from clinical research on the role and design of real-data benchmark studies

Clinical research Treated in Transfer into computational Example(s)
   statistical research?  
Sample size calculation [9] Possible and desirable [9] [35]
Strict inclusion criteria Sec. 3 Possible and desirable [20, 21, 35]
Trial protocol Sec. 4.1 Principle might be helpful in adapted form  
Quality control Sec. 4.2 Principle might be helpful in adapted form  
   e.g. via platforms like OpenML [4]  
Placebo/reference Sec. 4.3 Principle might be helpful in adapted form  
Blinding Sec. 4.4.1 Principle might be helpful in adapted form  
Intention-to-treat Sec. 4.4.2 Adequate treatment and reporting of [29]
   missing values: possible and desirable  
Levels of evidence Sec. 4.5 Principle might be helpful in adapted form