(a) Triggering. First of all, 99.99% of the data must be thrown away! So there needs to be a trigger to accept or reject” collision data for analysis–whether for immediate processing or for later on, as in so-called “data parking”.
With triggering we are not far off the idea that a result of a “test”, or single piece of data analysis, is to take one “action” or another:
reject the null -> retain the data;
do not reject -> discard the data.
(Here the null might, in effect, hypothesize that the data are not interesting.) It is an automatic classification scheme, given limits of processing and storing; the goal of controlling the rates of retaining uninteresting and discarding potentially interesting data is paramount.[i] It is common for performance oriented tasks to enter, especially in getting the data for analysis, and they too are very much under the error statistical umbrella.
Particle physicist Matt Strassler has excellent discussions of triggering and parking on his blog “Of Particular Significance”. Here’s just one passage:
Data Parking at CMS (and the Delayed Data Stream at ATLAS) takes advantage of the fact that the computing bottleneck for dealing with all this data is not data storage, but data processing. The experiments only have enough computing power to process about 300 – 400 bunch-crossings per second. But at some point the experimenters concluded that they could afford to store more than this, as long as they had time to process it later. That would never happen if the LHC were running continuously, because all the computers needed to process the stored data from the previous year would instead be needed to process the new data from the current year. But the 2013-2014 shutdown of the LHC, for repairs and for upgrading the energy from 8 TeV toward 14 TeV, allows for the following possibility: record and store extra data in 2012, but don’t process it until 2013, when there won’t be additional data coming in. It’s like catching more fish faster than you can possibly clean and cook them — a complete waste of effort — until you realize that summer’s coming to an end, and there’s a huge freezer next door in which you can store the extra fish until winter, when you won’t be fishing and will have time to process them.
(b) Bump indication. Then there are rules for identifying bumps, excesses more than 2 or 3 standard deviations above what is expected or predicted. This may be the typical single significance test serving as more of an indicator rule. Observed signals are classified as either rejecting, or failing to reject, a null hypothesis of “mere background”; non-null indications are bumps, deemed potentially interesting. Estimates of the magnitude of any departures are reported and graphically displayed. They are not merely searching for discrepancies with the “no Higgs particle” hypothesis, they are looking for discrepancies with the simplest type, the simple Standard Model Higgs. I discussed this in my first flukes post.
How much additional checking of assumptions, data analysis, etc.should be required before reporting a prima facie statistically significant effect? It varies. Knowing the discretionary standard used, we “consumers” can scrutinize them. We must also ask, report to whom? It appears that in experimental particle physics, at least in this case, they will internally report an indication of a statistically significant bump, but they are careful not to leak it even to the full physics community, at least if it seems to indicate some exotic particle at odds with the simple Standard Model Higgs. At least that is my impression from reading some of the literature. Thus far, all of the potentially exciting (anomalous) indications disappear with further data; the very thing that is expected were the anomalous effects mere flukes. This is not a flaw with the “bump indication” rule, at least not in an experimental particle physicist’s context, because it is known that these indications will be checked and cross-checked, that effects will have to stand up to severe scrutiny.
From indication to evidence
This takes us back to where I began (here), evidence of a 5 sigma effect, then, refers to the evidence after scrutiny of bumps will not go away, and to the 2013 data on many more collisions. The particle’s properties with respect to a given type of decay do not change; finding it again and again substantiates a strong “argument from coincidence” of the sort that severity requires.
But that is not all that is inferred, in my view. It is also crucial is to infer, and report on, what has not been well-probed or severely indicated. In particular, here, they have not distinguished various properties of the particle, and have not ruled out alternatives to the Standard Model. Knowing what has been poorly distinguished will surely be the basis for future design recommendations. Progress is not inferring a hypothesis or theory “out there” but the growing understanding of the phenomenon, as modeled, triggered, simulated, and probed.