Hello everybody! I am using the Curve Fitting Toolbox of Matlab to fit some non-linear models to my data, but I want to know which model fits better. I read that I have to use the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC) but I do not have those options in the Curve Fitting Toolbox, only R square, adjusted R square, SSE and RMSE. Do you know how can I get AIC and BIC for a model? Maybe without the Curve Fitting Toolbox? I would appreciate your help.
Hi Sergio,
If they do not appear in the Curve Fitting Toolbox, then you may compute the AIC and BIC values using the 'aicbic' command, which belongs to the Econometrics Toolbox.
https://www.mathworks.com/help/econ/aicbic.html
Dear Sergio Valdivia Rivera,
You can get valuable information on how to use AIC and BIC in MATLAB by doing the following:
1. Run your MATLAB
2. Go to the command window and type
help aic
then press enter key
it will display information about aic and examples on how it could be used.
you repeat the same procedure for bic by typing :
help bic
in your command line and press enter key.
I hope you would find this tips helpful.
Best wishes.
You can always use the adjusted R squared since it also takes into account the total number of model parameters.
The model with the lowest AIC has the best fit. Therefore, the fourth model fits best.
see this link
https://de.mathworks.com/help/econ/aicbic.html#btvzm7d-3
Hello everybody
Can i calculate AIC using excel?
I would appreciate your help
I recommend this routine
https://www.mathworks.com/matlabcentral/fileexchange/36000-fbd-find-the-best-distribution-tool
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Physical Phenomena
- Luminescence
- Fluorescence