How to estimate confidence intervals for model patamerters in Cadet match/Cadet

Dear All,
I am using Python based Cadet match to estimate the model parameters (Axial dispersion coefficient, Film mass transfer coefficient, Pore diffusion coeff. and surface diffusion coeff.) present in 1-D GRM.
Anyone please let me know how to get the confidence intervals for each model patamerter.

Your support in this regard will be highly appreciated.

CADETMatch has error modeling built into the system but it can be intensive to calculate.

You can see settings for this in Error Modeling — CADET-Match and Search — CADET-Match. This is not something that you can get from just running an optimization procedure.

1 Like

Does CADET-Match calculate CIs based on parameters sensitivities (jacobian → hessian → covariance → CIs)?

No, it does it using MCMC.

The approach you just outlined I am pretty sure can’t work for chromatography because the parameters are non-linear and coupled. It will violate the assumptions inherent in the process you outlined. You will get an answer but I am pretty sure it would always be wrong.

You also can’t use parameter sensitivities to calculate an optimal design for chromatography because optimal design theory has linearity assumptions built into it also. We tried that at one point doing my PhD and it just did not work out.

1 Like