ajgilbert / eft-exercise-cms Goto Github PK

Ultimate goal of the exercise is to go cross-experiment so certainly will need this.

Currently we're compiling Combine and CombineHarvester just for the purpose of plotting - would be better to have the relevant plotting code directly in this repository so we can avoid the additional compilation step.

Further to #7, this case is more complex because the PCA will change with phase space. Ideally it should be done at the best fit point in the WC space, but this is not straightforward to determine as there will be degeneracies. Perhaps we can think of a different way to handle this.

Needed to determine the combinations of WC's that can be constrained simultaneously. In the first instance, do this only for the linear parameterisation (more complicated for quadratic case)

Used when producing the scaling term json files in EFT2Obs

These are currently hardcoded in the fitting script, for more flexibility it would be better to read these from a separate config.

Currently only fitting in 1D with the other WC's fixed to 0; we should implement two-dimensional scans as well.

Would be nice to have a good way of visualising contributions from the different inputs to constraints on a specific WC.

Only including parameterisations for a subset of the STXS bins at the moment, would be nice to include others bins as well (nb: along the same lines, need to see what we can take from the work being done for the Higgs combination on these parameterisations)

See slide 10 here. That is an example of the yaml format for parameterisations, need to double check what is needed for the inputs.

This is to harmonize with ATLAS.

Would be useful to have options to drop measurement bins or parameterisations of specific bins easily. This would help with clipping studies.