Signal strength parameter (interpretation)

[ChrisVer]

How can in general the signal strength parameter $\mu$ be interpreted?

I am talking for the parameter defined in Eq.1 here and plots like the Fig.1 here:
http://arxiv.org/abs/1507.04548

It says that it's the ratio of the i->H->f of the observed over what's expected by the SM... is the last the cross section prediction of the Higgs or for any other background?
Then what would the <1 or >1 indicate? I think the >1 indicate a signal excess, while the <1 indicate a signal underestimation(?)

 

[mfb]

is the last the cross section prediction of the Higgs or for any other background?

Only Higgs. The background in data is subtracted before μ is calculated.

Then what would the <1 or >1 indicate?

A deviation from the standard model. If μ=1 gets ruled out in some channel, things get interesting.

 

[ChrisVer]

Only Higgs. The background in data is subtracted before μ is calculated.

Is that the case even if you have background overestimation compared to data?

 

[mfb]

I'm not sure if I understand your question. If you overestimate something, you are doing something wrong and should fix it, or not use what you cannot get right.

 

[ChrisVer]

I'm not sure if I understand your question. If you overestimate something, you are doing something wrong and should fix it, or not use what you cannot get right.

Well it's not unusual to get overestimates like these [at <120 or 140GeV]:
http://cms.web.cern.ch/sites/cms.we...ZMass_7Plus8TeV_70-180_3GeV.png?itok=k2MlxuLT
In which case subtracting the estimates from data will result to negative values.

 

[mfb]

That is (hopefully) not an overestimate, just a statistical fluctuation. Yes, estimated signal strengths can be negative. As a random example, this ATLAS note has -0.4 +- 1.1 for VH -> Vbb in table 2. Note that it is consistent with 1, the uncertainties were just very large in 2012.