So far, the Jacobian of ODEModels is determined Numerically. However, in fitting problems with many parameters, the fitting sometimes goes completely of the rails and gives up. I wonder if this could have something to do with not estimating the Jacobian properly, and wether is could be solved by providing it analytically.
So I should investigate again if it is possible to define this analytically.
tBuLi
commented
6 years ago
So far, the Jacobian of ODEModels is determined Numerically. However, in fitting problems with many parameters, the fitting sometimes goes completely of the rails and gives up. I wonder if this could have something to do with not estimating the Jacobian properly, and wether is could be solved by providing it analytically.
So I should investigate again if it is possible to define this analytically.