Since, just before I was going
to post this paper there has been the joint ATLAS and CMS
seminar on the Higgs boson search at CERN, followed by days of rumors,
I cannot avoid to add here some last minute comments
on these results, comparing them with the CDF case.
The big difference between the Fermilab result discussed
here and that of CERN[31] is essentially
a question of priors, whose role
was discussed in section 5.
If we observe something
unexpected, we need an overwhelming experimental evidence
before we are convinced this is really a genuine discovery,
which is not case of the highly expected Higgs at LHC.
These are the arguments in favor of the fact that
the elusive beast has been finally surrounded
(every particle hunter sniffs it, although
it will be considered to finally in `our hands' only when we
shall be able, with the increasing number of events, to
study its behavior, such as decay modes etc.):
- the so called Standard Model of particle physics
provides an excellent description of
a network of experimental facts, and such a particle
is required to give a sense to the theory;
- the indirect information on the Higgs boson (`radiative corrections')
constrains its mass at the order of magnitude of 100GeV
(although with a large uncertainty
- see [18] for a probability distribution,
even though this has been slightly changing with time);
- direct searches at LEP have pushed its mass with almost
certainty above 114GeV;17
- similarly, direct searches at the LHC and at the Tevatron
have squeezed its mass value into a relative
narrow window (I save the reader yet another disquisition
on the meaning of those limits);
- the CERN indication shows up
- in the middle of the remaining window of possibility
(and then not in contradiction with other experimental
pieces of information);
- with production rate in agreement with
the theory and with many other experiments
(from which the theoretical parameters have been
inferred);
- with decay modes also in substantial agreement
with expectations;
- in two detectors, although with some differences
that can be considered physiological,
taking into account of the difficulty of the search.
In addition, I would also like to remark that the presentations
of the two team leaders have been rather prudential,
as if, instead of the Higgs, it were
just an unexpected
bunch of extra events in the middle of nowhere.
Some further remarks are in order.
- The reason why practically
every particle physicist is highly confident that
the Higgs is in the region indicated by LHC
has little to do with the number of sigma's
(I hope the reader understands now that the mythical value of
5 for a `true discovery' is by itself pure nonsense,
as it is clear from the comparison between
`???' at the Tevatron
and the Higgs boson at CERN in the only place it could be
after it has been hunted unfruitfully
elsewhere.18)
- This number of sigma's cannot be turned in
probabilistic statements (or odds!)
about Higgs or not-Higgs,
as we read again on The New York Times:19
The Atlas result has a chance of less than one
part in 5,000 of being due to a lucky background noise,
which is impressive but far short of the standard
for a “discovery,†which requires one in 3.5 million
odds of being a random fluctuation.[30]
(Again misinterpreted p-values - basta!)
- Instead, if we want to make quantitative probabilistic assessments,
we need the likelihoods (this time this noun
has the technical meaning statisticians use), per each experiment
and per each channel, instead of the frequentistic
95% CL exclusion curves, of dubious meaning and useless
to be combined. A plea to the LHC collaborations is therefore in order:
please publish likelihoods.
- In the past days I have visited some internet resources
to check the rumors. As a result
Giulio D'Agostini
2012-01-02
