{"title":"点阵 QCD 中由引导法确定的 p 值","authors":"Norman Christ, Rajiv Eranki, Christopher Kelly","doi":"arxiv-2409.11379","DOIUrl":null,"url":null,"abstract":"We present a general method to determine the probability that stochastic\nMonte Carlo data, in particular those generated in a lattice QCD calculation,\nwould have been obtained were that data drawn from the distribution predicted\nby a given theoretical hypothesis. Such a probability, or p-value, is often\nused as an important heuristic measure of the validity of that hypothesis. The\nproposed method offers the benefit that it remains usable in cases where the\nstandard Hotelling $T^2$ methods based on the conventional $\\chi^2$ statistic\ndo not apply, such as for uncorrelated fits. Specifically, we analyze a general\nalternative to the correlated $\\chi^2$ statistic referred to as $q^2$, and show\nhow to use the bootstrap as a data-driven method to determine the expected\ndistribution of $q^2$ for a given hypothesis with minimal assumptions. This\ndistribution can then be used to determine the p-value for a fit to the data.\nWe also describe a bootstrap approach for quantifying the impact upon this\np-value of estimating population parameters from a single ensemble of $N$\nsamples. The overall method is accurate up to a $1/N$ bias which we do not\nattempt to quantify.","PeriodicalId":501191,"journal":{"name":"arXiv - PHYS - High Energy Physics - Lattice","volume":"39 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Bootstrap-determined p-values in Lattice QCD\",\"authors\":\"Norman Christ, Rajiv Eranki, Christopher Kelly\",\"doi\":\"arxiv-2409.11379\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"We present a general method to determine the probability that stochastic\\nMonte Carlo data, in particular those generated in a lattice QCD calculation,\\nwould have been obtained were that data drawn from the distribution predicted\\nby a given theoretical hypothesis. Such a probability, or p-value, is often\\nused as an important heuristic measure of the validity of that hypothesis. The\\nproposed method offers the benefit that it remains usable in cases where the\\nstandard Hotelling $T^2$ methods based on the conventional $\\\\chi^2$ statistic\\ndo not apply, such as for uncorrelated fits. Specifically, we analyze a general\\nalternative to the correlated $\\\\chi^2$ statistic referred to as $q^2$, and show\\nhow to use the bootstrap as a data-driven method to determine the expected\\ndistribution of $q^2$ for a given hypothesis with minimal assumptions. This\\ndistribution can then be used to determine the p-value for a fit to the data.\\nWe also describe a bootstrap approach for quantifying the impact upon this\\np-value of estimating population parameters from a single ensemble of $N$\\nsamples. The overall method is accurate up to a $1/N$ bias which we do not\\nattempt to quantify.\",\"PeriodicalId\":501191,\"journal\":{\"name\":\"arXiv - PHYS - High Energy Physics - Lattice\",\"volume\":\"39 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-17\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - PHYS - High Energy Physics - Lattice\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.11379\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - PHYS - High Energy Physics - Lattice","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.11379","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
我们提出了一种通用方法来确定随机蒙特卡洛数据的概率,特别是在格子 QCD 计算中生成的数据,如果这些数据来自给定理论假设所预测的分布,那么这些数据就会被得到。这种概率或 p 值经常被用作衡量假设有效性的重要启发式指标。拟议方法的好处是,在基于传统$\chi^2$统计量的标准霍特林$T^2$方法不适用的情况下,例如在不相关拟合的情况下,它仍然可用。具体来说,我们分析了相关$\chi^2$统计量的一般替代方法,称为$q^2$,并展示了如何使用引导法作为数据驱动方法,以最小的假设来确定给定假设下$q^2$的预期分布。我们还介绍了一种自举法,用于量化从 $N$ 样本的单一集合中估计群体参数对 p 值的影响。整个方法的精确度可达到 1/N$ 的偏差,但我们并不试图对其进行量化。
We present a general method to determine the probability that stochastic
Monte Carlo data, in particular those generated in a lattice QCD calculation,
would have been obtained were that data drawn from the distribution predicted
by a given theoretical hypothesis. Such a probability, or p-value, is often
used as an important heuristic measure of the validity of that hypothesis. The
proposed method offers the benefit that it remains usable in cases where the
standard Hotelling $T^2$ methods based on the conventional $\chi^2$ statistic
do not apply, such as for uncorrelated fits. Specifically, we analyze a general
alternative to the correlated $\chi^2$ statistic referred to as $q^2$, and show
how to use the bootstrap as a data-driven method to determine the expected
distribution of $q^2$ for a given hypothesis with minimal assumptions. This
distribution can then be used to determine the p-value for a fit to the data.
We also describe a bootstrap approach for quantifying the impact upon this
p-value of estimating population parameters from a single ensemble of $N$
samples. The overall method is accurate up to a $1/N$ bias which we do not
attempt to quantify.