AIC model comparison
Rank likelihood fits and inspect their relative support within a candidate set.
Your entries stay in this page’s memory and are not submitted to us. Downloads are generated locally. Reloading or leaving the page clears your entries. Do not enter patient information.
Enter at least two models, each with a unique label, maximized natural-log likelihood and total estimated parameter count. Include variance parameters where estimated. Lower AIC ranks first; delta is the difference from the lowest AIC. Weights sum to one across only the models entered and are not probabilities that a model is scientifically true.
The starting values are hypothetical. Results update as you edit. Use aggregate or synthetic inputs only.
Calculation results
- Lowest AIC
- 200
- Ranking
- Extended (Δ 0, weight 0.8808) · Base (Δ 4, weight 0.1192)
How to interpret the output
AIC equals 2k − 2 log likelihood. Include all estimated parameters. Compare maximum-likelihood fits to the same observations and outcome using comparable likelihood constants. Relative weights depend on the candidate set.
Method reference