1 |
|
2 |
\subsection{Single Lepton Top MC Modelling Validation from CR2}
|
3 |
\label{sec:cr2}
|
4 |
|
5 |
|
6 |
The \mt\ tail for single-lepton top events (\ttsl\ and single top) is dominated by jet resolution effects. The \W\ cannot be far off-shell because $\mW < \mtop$.
|
7 |
The modeling of the \mt\ tail from jet resolution effects can be studied
|
8 |
using \zjets\ data and MC samples. However, as we will show below,
|
9 |
this test is statistically limited and can only be performed for the
|
10 |
\met\ requirements corresponding to SRA and SRB.
|
11 |
|
12 |
\Z\ events are selected by requiring exactly 2 good leptons (satisfying ID
|
13 |
and isolation requirements) and requiring the \mll\ to be in the range
|
14 |
$81-101$ GeV.
|
15 |
Events with additional isolated tracks are vetoed, as in Section~\ref{sec:tkveto}.
|
16 |
To reduce \ttbar\ backgrounds, events with a CSVM tag %H
|
17 |
are removed.
|
18 |
The positive lepton is treated as a neutrino and so is added to the MET: \met\ $\rightarrow$ \pt(\Lepp) + \met,
|
19 |
and the \mt\ is recalculated with the negative lepton: \mt(\Lepm, \met).
|
20 |
The resulting ``pseudo-\mt'' is dominated by jet resolution effects, since no off-shell
|
21 |
\Z\ production enters the sample due to the \mll\ requirement.
|
22 |
This section describes how well the MC predicts the tail of ``pseudo-\mt''.
|
23 |
|
24 |
The underlying distributions are shown in Fig.~\ref{fig:cr2met}.
|
25 |
%and~\ref{fig:cr2mtrest}.
|
26 |
We then perform the exact same type of Data/MC comparison and analysis as
|
27 |
described for CR1 in Section~\ref{sec:cr1}. For CR1 we collected
|
28 |
the data/MC tail information in
|
29 |
Table~\ref{tab:cr1yields}; the equivalent for CR2 is
|
30 |
Table~\ref{tab:cr2yields} (for CR2 the statistics are not sufficient to split electrons and muons).
|
31 |
The last line of Table~\ref{tab:cr2yields} gives the data/MC scale factors
|
32 |
for the \ttbar\ lepton $+$ jets $M_T$ tail ($SFR_{top}$). This is
|
33 |
calculated in the same way as $SFR_{wjets}$ of Table~\ref{tab:cr1yields}.
|
34 |
Just as in CR1, there is an excess of data in the tails, as reflected
|
35 |
in the values of $SFR_{top}$. There are insufficient events to derive scale factors for
|
36 |
$\met\ > 150$~GeV. As a result, the scale factors derived from CR2 are
|
37 |
not used for the central prediction of the single-lepton top
|
38 |
background. They serve as a valuable cross check of the predictions
|
39 |
described in Section~\ref{sec:ttp}. The single lepton top predictions
|
40 |
obtained for SRA and SRB using the $SFR_{top}$ values described here
|
41 |
are consistent with the default predictions.
|
42 |
|
43 |
|
44 |
\begin{table}[!h]
|
45 |
\begin{center}
|
46 |
{\footnotesize
|
47 |
\begin{tabular}{l||c|c||c|c}
|
48 |
\hline
|
49 |
Sample & CR2PRESEL0 &CR2PRESEL1 & CR2A & CR2B \\
|
50 |
\hline
|
51 |
\hline
|
52 |
MC & $32 \pm 2$ & $28 \pm 2$ & $10 \pm 1$ & $10 \pm 1$ \\
|
53 |
Data & $50$ & $45$ & $17$ & $17$ \\
|
54 |
\hline
|
55 |
Data/MC & $1.56 \pm 0.24$ & $1.63 \pm 0.27$ & $1.68 \pm 0.45$ & $1.74 \pm 0.48$ \\
|
56 |
\hline
|
57 |
\hline
|
58 |
\hline
|
59 |
DY MC & $25 \pm 2$ & $20 \pm 2$ & $5 \pm 1$ & $5 \pm 1$ \\
|
60 |
DY Data & $42 \pm 7$ & $38 \pm 7$ & $12 \pm 4$ & $12 \pm 4$ \\
|
61 |
\hline
|
62 |
DY Data/MC & $1.73 \pm 0.32$ & $1.85 \pm 0.37$ & $2.37 \pm 0.96$ & $2.58 \pm 1.16$ \\
|
63 |
\hline
|
64 |
\hline
|
65 |
\hline
|
66 |
$SFR_{top}$ & $1.64 \pm 0.40$ & $1.74 \pm 0.46$ & $2.02 \pm 0.68$ & $2.16 \pm 0.75$ \\
|
67 |
\hline
|
68 |
\end{tabular}}
|
69 |
\caption{ Yields in \mt\ tail comparing the \zjets\ MC prediction (after
|
70 |
applying SFs) to data without subtracting the non-\zjets\ components (top table) and with subtracting the non-\zjets\ components (bottom table).
|
71 |
CR2PRESEL refers to a sample with $\met>50$ GeV and $\mt>150$ GeV.
|
72 |
\label{tab:cr2yields}}
|
73 |
\end{center}
|
74 |
\end{table}
|
75 |
|
76 |
%\hline
|
77 |
%$N_{1l-top}$ SF & - & - & $172 \pm 58$ & $119 \pm 42$ \\
|
78 |
%\hline
|
79 |
%$N_{1l-top}$ Opt/Pess & - & - & $256 \pm 131$ & $120 \pm 50$ \\
|
80 |
|
81 |
|
82 |
\begin{figure}[hbt]
|
83 |
\begin{center}
|
84 |
% \includegraphics[width=0.5\linewidth]{plots/CR2plots/met_scaled_nj4_emucomb.pdf}%
|
85 |
\includegraphics[width=0.5\linewidth]{plots/CR2plots/met_lepcor_scaled_nj4_emucomb.pdf}%
|
86 |
\includegraphics[width=0.5\linewidth]{plots/CR2plots/mt_lepcor_scaled_nj4_emucomb.pdf}
|
87 |
\includegraphics[width=0.5\linewidth]{plots/CR2plots/mt_lepcor_scaled_met50_nj4_emucomb.pdf}%
|
88 |
\includegraphics[width=0.5\linewidth]{plots/CR2plots/mt_lepcor_scaled_met100_nj4_emucomb.pdf}
|
89 |
|
90 |
\caption{
|
91 |
Comparison of the pseudo-\met\ (top, left), pseudo-\mt\ (top,
|
92 |
right and bottom) distributions in data vs. MC for events
|
93 |
satisfying the requirements of CR2, combining both the muon and
|
94 |
electron channels. The pseudo-\mt\ distributions are shown
|
95 |
before any additional requirements (top, right) and after
|
96 |
requiring pseudo-\met $>$50 GeV (bottom, left) and pseudo-\met
|
97 |
$>$ 100 GeV (bottom, right).
|
98 |
\label{fig:cr2met}
|
99 |
}
|
100 |
\end{center}
|
101 |
\end{figure}
|
102 |
|
103 |
%\begin{figure}[hbt]
|
104 |
% \begin{center}
|
105 |
% \includegraphics[width=0.5\linewidth]{plots/CR2plots/mt_lepcor_scaled_met150_nj4_emucomb.pdf}%
|
106 |
% \includegraphics[width=0.5\linewidth]{plots/CR2plots/mt_lepcor_scaled_met200_nj4_emucomb.pdf}
|
107 |
% \includegraphics[width=0.5\linewidth]{plots/CR2plots/mt_lepcor_scaled_met250_nj4_emucomb.pdf}%
|
108 |
% \includegraphics[width=0.5\linewidth]{plots/CR2plots/mt_lepcor_scaled_met300_nj4_emucomb.pdf}
|
109 |
% \caption{
|
110 |
% Comparison of the \mt\ distribution in data vs. MC for events
|
111 |
% satisfying the requirements of CR2, combining both the muon and
|
112 |
% electron channels. The pseudo-\met\ requirements used are
|
113 |
% 150 GeV (top, left), 200 GeV (top, right), 250 GeV (bottom,
|
114 |
% left) and 300 GeV (bottom, right).
|
115 |
%\label{fig:cr2mtrest}
|
116 |
%}
|
117 |
% \end{center}
|
118 |
%\end{figure}
|
119 |
\clearpage
|