Can I impose monotonicity constrain in my experiment?

We do not support monotonicity constrains at, because it enforces external assumptions into the model rather then letting the model estimate the coefficient based on the data.

In some cases, imposing such constraint may work well, but it will be difficult for us to implement this into our automated solutions as it will raise validity concerns on aggregate results (all our default outputs and simulation results). You are of course, more than welcome to export the raw data and re-estimate the model by imposing monotonicity constraint.