Skip to content
Prev 109466 / 398500 Next

how to report logistic regression results

Edwin Commandeur wrote:
There are many ways, including odds ratios and partial effect plots and 
Brier scores.  For a pure likelihood measure I talk about an 'adequacy 
index' (adequacy of the smaller model) in my book, which was used in a 
medical paper:

@ARTICLE{cal85,
   author = {Califf, R. M. and Phillips, H. R. and others},
   year = 1985,
   title = {Prognostic value of a coronary artery jeopardy score},
   journal = J Am Coll Cardiology,
   volume = 5,
   pages = {1055-1063}
}

Frank Harrell