"Table Aux. Figure 11 Medium-$D_\zeta$ 2016 Low-mass" of "Searches for additional Higgs bosons and for vector leptoquarks in $\tau\tau$ final states in proton-proton collisions at $\sqrt{s}$ = 13 TeV"

Observed and expected distributions of the variable chosen for statistical inference in the low-mass analysis $m_{\tau\tau}$. Numerical values provided in this table correspond to Figure 11 of the auxilliary material of the publication, but restricted to Medium-$D_\zeta$ category and 2016 data-taking year. All distributions are considered after a fit to data is performed using a background-only model, which includes the $\text{H}(125)$ boson. Some details on how the distributions should be used:

1) All given uncertainties correspond to systematic variations of $\pm1\sigma$.
2) Upper values ('plus' in the yaml file) correspond to an upward systematic variation of the parameter ($+1\sigma$).
3) Lower values ('minus' in the yaml file) correspond to a downward systematic variation of the parameter ($-1\sigma$).
4) These variations can have both positive and negative values, depending on the modelled effect.
5) Uncertainties with the same name should be treated as correlated, consistently across the upper and lower variations.
6) Systematic uncertainties with 'prop_' in the name treat limited background statistics per histogram bin, and are deployed with 'Barlow-Beeston-lite' approach. Details in https://arxiv.org/abs/1103.0354 section 5
7) Remaining systematic uncertainties alter the normalization, the shape, or both for a distribution. The nuisance parameter for such an uncertainty is mapped separately on the normalization and the shape variation components of the uncertainty. For normalization, $\ln$ mapping is used, for shape a spline. Details in https://cms-analysis.github.io/HiggsAnalysis-CombinedLimit/part2/settinguptheanalysis/#binned-shape-analysis
8) All nuisance parameters for the systematic uncertainties are modelled with a Gaussian pdf.
9) Gluon fusion contributions are all scaled to 1 pb. Please combine them using either the scale factors from 'Table SM Gluon Fusion Fractions', or using your own composition.

低质量分析中用于统计推断的变量$m_{ au au}$的观测分布与预期分布。本表提供的数值对应于该出版物辅助材料中的图11，但仅限于中等$D_zeta$类别和2016年数据采集年份。所有分布均为使用仅背景模型对数据进行拟合后得到的结果，该模型包含$ ext{H}(125)$玻色子。关于分布使用方式的若干细节：

1) 所有给出的不确定性对应于$pm1sigma$的系统偏差。
2) 上限值（YAML文件中的'plus'）对应于参数的向上系统偏差（$+1sigma$）。
3) 下限值（YAML文件中的'minus'）对应于参数的向下系统偏差（$-1sigma$）。
4) 这些偏差可根据建模效应取正值或负值。
5) 名称相同的不确定性应视为相关，且在上限与下限偏差中保持一致。
6) 名称中带有'prop_'的系统不确定性处理每个直方图bin的有限背景统计量，并采用'Barlow-Beeston-lite'方法。详情见https://arxiv.org/abs/1103.0354第5节
7) 其余系统不确定性会改变分布的归一化、形状或两者兼而有之。此类不确定性对应的nuisance参数（nuisance parameter）分别映射到不确定性的归一化和形状偏差分量上。归一化采用对数映射，形状采用样条。详情见https://cms-analysis.github.io/HiggsAnalysis-CombinedLimit/part2/settinguptheanalysis/#binned-shape-analysis
8) 系统不确定性的所有nuisance参数均采用高斯概率密度函数（Gaussian probability density function）建模。
9) 胶子聚变贡献均已缩放至1 pb。请使用'SM胶子聚变分数表'中的缩放因子或将其与您自己的组成结合使用。