jump to navigation

New Higgs limit combination from CDF! January 15, 2008

Posted by dorigo in news, physics, science.

Today my colleague Wei-ming Yao is showing the new combined limits on Higgs boson production by CDF at the Aspen winter conference. The result has been blessed just a week ago, and it is the result of the combination of several different analyses searching the Higgs boson in all the final states the Tevatron is sensitive to.

The data analyzed varies from 1.1 to 1.9 inverse femtobarns, and is thus only about a half of what has so far been collected by CDF. That is because these analyses are really difficult to perform, and the required accuracy demands a periodic re-tuning of all the devilish details – which takes a huge effort from the physicists involved in the analyses.

The limit is an update of the combination of results CDF published last summer, which I discussed back then in another post. The progress from the former numbers is small but important: in particular, a significant decrease of the cross section limit can be observed at low Higgs masses – the region where associated production of WH or ZH pairs contributes the most. Associated production of a vector boson and the Higgs is a tough signal to search for, because one needs to refine two of the most complex tools at a hadron collider: b-quark tagging -which allows to select the H \to b \bar b decay- and jet energy resolution, which affects the reconstruction of a mass bump.

The plot below summarizes the current state of affairs: as by now every reader of this blog should be familiar with, the x axis in this plot is the unknown value of the Higgs mass. On the y axis is the ratio between observed limit on higgs production cross section and standard model predictions for the same process: a limit at y=2.0 means that the Higgs has been determined to not be produced at rates exceeding twice those predicted by Standard Model.

You can see several lines in the plot. Each color corresponds to a different search: red stands for the WH \to l \nu b \bar b associated production search, while blue stands for the analysis of ZH \to \nu \nu b \bar b decays when the Z goes to neutrinos; in purple, instead, are the results of a search for the ZH \to l l b \bar b  final state. Finally, the green lines detail the results of the search for direct H production and decay to a pair of W bosons.

For each search, the two lines have a different meaning: the hatched line is the limit CDF expected to set, given the analysis details and the amount of analyzed data; the full line shows instead the limit actually obtained. The differences show where CDF has been “lucky” (if one may claim to be lucky when observing fewer events than background processes should give) or unlucky.

What matters most, however, is the black line, which is the combined limit, obtained from a complex statistical procedure on which I have no desire to delve right now. The yellow band shows the variability of the expected combined limit.

What should we take home from this plot ? Well, several things. First, we see that the addition of 70% more data has made the ZH \to \nu \nu b \bar b search the most sensitive at very low masses. This brought down the limit from last summer’s x10 at Mh=120 GeV to today’s x5.5. Help came also from the WH search, which has added about 15% more data since last summer.

All in all, the CDF limit stands at x1.8 times the Standard Model for higgs boson mass of 160 GeV. I think we will have to wait for summer 2008 to see the first chunk of Higgs masses being excluded right about there by a combination of CDF and D0 results.


1. Myke - January 16, 2008

Is it just me or do others think this is a rather non-intuitive way to present the results?


2. dorigo - January 16, 2008

Well Myke, it is indeed a bit non-intuitive, but once you get a grip of it it becomes a very useful way to compare the limit at different mass points. Otherwise, when you place the limit in the cross section, you have to compare it to the steeply falling SM curve, and you find yourself taking mentally the ration.


3. Kea - January 16, 2008

Fun, fun, fun! Glad some physicists get to enjoy Aspen.

4. dorigo - January 17, 2008

Yeah Kea, but not me unfortunately! For some reason I never got chosen for that conference. The CDF Speakers Committee allocates talks to members of the collaboration, and some of the nicest spots are always taken by others 😦
Of course, I cannot complain, because I did attend nice conferences in the past. If you propose yourself for a conference, and are not too picky, you can represent CDF about once a year somewhere. And the places farthest from the US are usually easier to get, unless we talk about ICHEP or ski conferences in the italian alps…


5. strings - January 22, 2008

It used to be claimed that with 2 fb^{-1} of data, the Tevatron
could exclude a Standard Model Higgs particle in the low
mass region near 115 GeV. It seems to me that as recently as a year
ago, Dorigo wrote that in his judgment, this estimate will hold up.

When one looks at the current plots, it is clear that with over 1 fb^{-1}, one is far from the sensitivity needed to exclude the Higgs particle in the low mass region.

Are the old estimates still applicable, or is there some reason that the Tevatron Higgs search lacks the sensitivity that was hoped for?

6. dorigo - January 22, 2008

Dear strings,

the relevant post from last May is https://dorigo.wordpress.com/2007/05/03/the-2003-higgs-discovery-predictions-tested-with-hard-data/
but of course I discussed the murky matter of limit-setting in several places on this blog in the last couple of years.

Quoting from last May’s post, I wrote:
The comparison is quite interesting indeed. We in fact find that at 100 GeV Dzero is doing 2.8 times worse than expected in 2003. At 140 GeV, it is doing 2.1 times worse. At 160 GeV, it is doing 1.4 times worse. And at 200 GeV, it is doing just 1.1 times worse than what one would extrapolate from the 1999/2003 study.
Now remember what we discussed above: analyses are tough, especially at low mass. We are, indeed, quite deep into Run II as far as data taking is concerned, but on the other hand analyses are only just starting to produce results close to the possible optimum. Recall that the best results on the top quark mass in Run I were produced by CDF and Dzero six years after the end of data taking, and the error shrunk by a factor of two from the earlier results based on the same dataset.
What that means, is that the Tevatron experiments are like a good wine: they get better with time. So, is a factor of 1.5 – 2.0 worse than expectation worrysome ? Does it mean that the 2003 predictions were far off the mark ?
My personal answer is a resounding NO. I am confident that CDF and Dzero will deliver results in line with expectations. And since I believe that the integrated luminosity delivered by the end of 2009 will most likely be around 7 inverse femtobarn per experiment (or so I extrapolate – will discuss that in another post soon), I think I can make some predictions here: a 115 GeV Higgs will produce 2.5 to 3.0 sigma evidence in Tevatron data if it is there. If not, an exclusion will be drawn up to 130 GeV.

Now, you see: the point is that we do not just need 2/fb. We also need a long time to refine our tools!

In any case a new look at the numbers is useful. CDF is at about 5xSM for a Higgs of 115 GeV in the plot above (please look at the hatched line, the expected limit – the one which is not sensitive to fluctuations in the actual data collected so far).
The luminosity of the searches that most contribute to the limit at 115GeV is 1, 1.7, and 1.9 femtobarns for the three channels pictured. This indeed means more than 1/fb, and if we look at the individual sensitivities of the three searches for associated production we’d have to concur the effective average luminosity used is of the order of 1.75/fb.

Other things to note are that the CDF search is a trifle less sensitive in this region than the D0 search – for this, compare to https://dorigo.wordpress.com/2007/12/18/new-d0-higgs-limit-combination-for-17fb/ , where you see that about the same sensitivity -5.5xSM – was achieved with low-mass searches using 0.9 to 1.1/fb in D0. Extrapolating those figures to 2/fb instead than those of CDF, which is ahead in statistics analyzed but behind in the refining of methods, one would get a 3.8xSM limit from D0 alone with 2/fb.

These numbers still suffer from the lack of combination with less sensitive analyses, which lag behind, from still not optimized mass reconstructions -a big improvement can be expected there once D0 improves their mass resolution- and to some extent to other tunings. Because of that, I think it is a tad unfair to conclude that the 2003 studies were way off the mark. We need time. The first combination of CDF and D0 Higgs analyses using 2/fb will not get close to excluding a 115 GeV Higgs, but they will be very close to exclude it at 160 GeV – where analyses are much more straightforward to carry out. In two more years, a refining of the analyses could bring limits down by as much as a factor of two, in my opinion.


7. Not Even Wrong » Blog Archive » Science, and a Bit of Religion - January 22, 2008

[…] a prediction of the Higgs mass, as twice the W-mass, about 160 Gev. Coincidentally, Tommaso Dorigo reports on the Higgs search at the Tevatron, which is getting close to being able to exclude the existence […]

Sorry comments are closed for this entry

%d bloggers like this: