I believe the order of the shape parameters is reversed. The help page for Gompertz-Makeham distribution states that the hazard function is defined as $h(x) = shape[1] + shape[2] \exp(x/scale)$ whereas throughout the code it is shape[2] + shape[1] \exp(x/scale).
Given that existing code may depend on this behavior, I believe it will be better to update the documentation rather than the code.
Hello, thank you for this package.
I believe the order of the
shape
parameters is reversed. The help page for Gompertz-Makeham distribution states that the hazard function is defined as $h(x) = shape[1] + shape[2] \exp(x/scale)$ whereas throughout the code it isshape[2] + shape[1] \exp(x/scale)
.Given that existing code may depend on this behavior, I believe it will be better to update the documentation rather than the code.
Best regards, Oleksandr