Statistical inference¶
The final step turns the merged histograms into datacards and runs limits with
Combine, via the StatInference and
inference submodules. See
FLAF → walkthrough, stage 5
for where this sits in the pipeline.
These commands run inside CMSSW/Combine, so prefix them with cmsEnv (or open one subshell):
1. Create datacards¶
cmsEnv python3 StatInference/dc_make/create_datacards.py \
--input PATH_TO_SHAPES \
--output PATH_TO_CARDS \
--config StatInference/config/x_hh_bbww_run3.yaml
The HH→bb̄WW Run 3 configuration is
x_hh_bbww_run3.yaml.
2. Run limits¶
Hints:
- add
--workflow htcondorto submit to the batch system (local by default); - add
--remove-output 4,a,yto clear previous outputs; - add
--print-status 0to get the workflow status and the output file name; - options and background: the cms-hh inference documentation.
3. Pulls & impacts¶
PlotPullsAndImpacts --version dev --datacards "PATH_TO_CARDS/<one_card>.txt" \
--hh-model NO_STR --parameter-values r=1 --parameter-ranges r,-100,100 \
--method robust --PlotPullsAndImpacts-order-by-impact True --mc-stats True \
--PullsAndImpacts-custom-args="--expectSignal=1"
One mass point at a time
Run pulls & impacts on a single datacard, not a glob. Use --print-status 0 to find the
output file and --remove-output 4,a,y to clear previous outputs.