14. Logistic Regression Part 4

Introduction to Statistical Modelling
With Dr Helen Brown, Senior Statistician at The Roslin Institute, December 2015
*Recommended Youtube playback settings for the best viewing experience: 1080p HD
************************************************
Content:
Logistic regression for prediction: General points
-Model performance:
--- Test predictions using another dataset
--- Or use ‘cross-validation’ (use subsets of data in turn for ‘training’ and evaluating on)
-Take care interpreting results for individual variables if many variables fitted
---Less meaningful when adjusted for many variables
---Adjusted and unadjusted results for PVD (peripheral vascular disease) :-
-Note
---Odds ratios with 95% CIs which overlap 1 will not be significant
---Other modelling methods also available for prediction and may sometimes perform better