Closed lhdp0110 closed 1 year ago
I'm not completely sure what you need/want, and you're not using symfit. It sounds like you want the jacobian (derivative wrt parameters) of your "analysis" functions. You can determine this either analytically, or by using something like finite difference methods.
I need to perform a sensitivity analysis measuring the impact of the model's parameters onto "indirect" aspects of the model's fit. By indirect aspects, I mean how the parameter variations affect the initial value, maximum value, induction rate (upwards slope from initial to max), and decay rate (downwards slope from max back to baseline) of the fit. I have written functions to quantify these four aspects, but cannot figure out how to perform a sensitivity analysis that is not directly of the function. Please help?