Statistical evaluation of fitted models applied to $\bf p_t$ spectrum data from 5 TeV and 13 TeV $\bf p$-$\bf p$ collisions at the large hadron collider

Abstract

In a recent analysis of high-statistics $p_t$ spectra from 5 and 13 TeV $p$-$p$ collisions a two-component (soft+hard) model (TCM) of hadron production near midrapidity, based on fixed model functions independent of collision multiplicity $n_{ch}$ or event selection criteria, was employed as a reference to determine data biases induced by the selection criteria. As in previous studies the fixed TCM accurately isolated jet-related and nonjet contributions to spectra. It was observed that two selection criteria (V0M and SPD), based on different pseudorapidity $\eta$ acceptances, bias the spectrum hard component (jet fragment distribution) in different but complementary ways whereas the soft component is not significantly biased. In the present study TCM model functions are adjusted (i.e. fitted) to accommodate data. The effect of selection bias is then represented by smooth evolution of model parameters with $n_{ch}$. To evaluate the quality of model fits, Z-scores (data-model differences divided by statistical uncertainties) are evaluated and compared with data/model ratios that are often used for such comparisons. Based on Z-scores the variable TCM is compared with two other frequently-invoked models: the Tsallis model and the blast-wave model as they have been recently applied to the same spectrum data. The results are relevant to recent claims that collectivity (various flow manifestations) as a possible manifestation of QGP formation is observed in small systems. Data systematics suggest that these $p$-$p$ $p_t$ spectra are consistent with conventional QCD.