Parameter sensitivity for each observation group?

[sjepsen395]

Hi All,

Does anyone know of PEST output containing parameter sensitivity for each group of observations? The sensitivities in the .sen file from PEST are summed over all rows of the Jacobian, giving parameter sensitivities relative to all observations. I'm looking for values summed only over rows of each observation group, giving sensitivity of parameter i to observation group j. I thought I had seen this output somewhere, but can't find it now. This would help inform parameter tying (manual regularization).

Thanks,
Steve