| 1 |
\section{Definition of the signal region}
|
| 2 |
\label{sec:sigregion}
|
| 3 |
|
| 4 |
We define a signal region to look for possible
|
| 5 |
new physics contributions in the opposite sign isolated
|
| 6 |
dilepton sample. The choice of signal region is driven by
|
| 7 |
three observations:
|
| 8 |
\begin{enumerate}
|
| 9 |
\item astrophisical evidence for dark matter suggests that
|
| 10 |
we concentrate on the region of high \met;
|
| 11 |
\item new physics signals should have high $\sqrt{\hat{s}}$;
|
| 12 |
\item observable high cross section new physics signals
|
| 13 |
are likely to be produced strongly; thus, we expect significant
|
| 14 |
hadronic activity in conjunction with the two leptons.
|
| 15 |
\end{enumerate}
|
| 16 |
|
| 17 |
Following these observations, we add the following two requirements
|
| 18 |
to the preselection of Section~\ref{sec:eventSel}:
|
| 19 |
\begin{center}
|
| 20 |
SumJetPt$>$300 GeV and $\met/\sqrt{\rm SumJetPt} > 8.5$.
|
| 21 |
\end{center}
|
| 22 |
|
| 23 |
\noindent This selection preserves about 1\% of the $t\bar{t}$
|
| 24 |
signal. We cut on \met$/\sqrt{\rm SumJetPt}$ rather than \met
|
| 25 |
because the variables \met and \met$/\sqrt{\rm SumJetPt}$ are
|
| 26 |
largely uncorrelated for the dominant $t\bar{t}$ background.
|
| 27 |
This allows us to use a data driven ABCD method to estimate the
|
| 28 |
background (see Section~\ref{sec:abcd}).
|
