{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Binned vs. unbinned fits\n", "\n", "We compare binned with unbinned fit applied to a toy example of a gaussian signal peak over exponential background.\n", "\n", "For one-dimensional data, binned fit are preferred. They are usually considerably faster than an unbinned fit and more numerically stable. For multi-dimensional data, however, an unbinned fit can be faster.\n", "\n", "It is a common misconception that binned fits are inherently biased. This idea originates from the past when it was common (at least in particle physics) to fit binned data with the least-squares method, which is indeed biased, see [Dembinski, Schmelling, Waldi, *Application of the Iterated Weighted Least-Squares Fit to counting experiments*, NIM A 940 (2019) 135-141](https://doi.org/10.1016/j.nima.2019.05.086). That bias can be completely avoided, however, if the fit uses the maximum-likelihood method and a Poisson distribution to describe the observed bin contents as a function of the predicted ones, and if the model prediction for a bin content is properly computed by integrating over the model density, instead of computing it from the density at the bin center times the bin width. The cost functions `BinnedNLL` and `ExtendedBinnedNLL` from `iminuit.cost` use the correct calculation.\n", "\n", "So there is no need to worry bias, but some information is lost in the binning process - the densities of events inside each bin. This loss can be made negligible by making the bin width small enough. How small the bins have to be depends on the sensitivity of the model parameter on this particular loss of information. In this tutorial we demonstrate this and also demonstrate the difference in run-time of unbinned and binned fits.\n", "\n", "**Conclusions:** With only 20 bins, the binned fit reached an accuracy for the signal yield that is comparable to the unbinned fit. With 50 bins, also all shape parameters have uncertainties that are less than 5 % larger than those in the unbinned fit. At the same time, the binned fit is much faster. Even with 200 bins, the binned fit is two orders of magnitude faster than the unbinned fit. In practice, this is a huge difference, 3 seconds vs. 5 minutes.\n", "\n", "You can try to run this notebook with a data sample contains less points, then the difference will not be as dramatic." ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [], "source": [ "import numpy as np\n", "from numba_stats import norm, expon\n", "import matplotlib.pyplot as plt\n", "from iminuit import Minuit\n", "from iminuit.cost import ExtendedUnbinnedNLL, ExtendedBinnedNLL\n", "import joblib\n", "from IPython.display import display" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [], "source": [ "# sample size, change this to see how the results of the comparison change\n", "n = 100_000\n", "truth = np.array((1.0, 1.0, 1.0, 0.1, 1.0))\n", "\n", "rng = np.random.default_rng(1)\n", "s = rng.normal(truth[2], truth[3], size=int(n * truth[0]))\n", "b = rng.exponential(truth[4], size=int(n * truth[1]))\n", "pts = np.append(s, b)\n", "pts = pts[(pts > 0) & (pts < 2)]" ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [], "source": [ "def fit(c):\n", " m = Minuit(c, s=1, b=1, mu=1, sigma=0.1, tau=1)\n", " m.limits[\"s\", \"b\", \"sigma\", \"tau\"] = (0, None)\n", " m.limits[\"mu\"] = (0, 2)\n", " m.migrad()\n", " assert m.valid\n", " return m" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Migrad
FCN = -4.08e+06 Nfcn = 120
EDM = 1.25e-05 (Goal: 0.0002) time = 0.9 sec
Valid Minimum Below EDM threshold (goal x 10)
No parameters at limit Below call limit
Hesse ok Covariance accurate
\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Name Value Hesse Error Minos Error- Minos Error+ Limit- Limit+ Fixed
0 s 0.995 0.004 0
1 b 1.008 0.005 0
2 mu 999.3e-3 0.4e-3 0 2
3 sigma 99.21e-3 0.34e-3 0
4 tau 1.002 0.007 0
\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
s b mu sigma tau
s 1.32e-05 -0.005e-3 (-0.281) -0.04e-6 (-0.029) 0.30e-6 (0.246) -0.004e-3 (-0.159)
b -0.005e-3 (-0.281) 2.36e-05 0 -0.46e-6 (-0.283) 0.020e-3 (0.586)
mu -0.04e-6 (-0.029) 0 1.46e-07 -0.01e-6 (-0.044) -0.14e-6 (-0.054)
sigma 0.30e-6 (0.246) -0.46e-6 (-0.283) -0.01e-6 (-0.044) 1.12e-07 -0.36e-6 (-0.157)
tau -0.004e-3 (-0.159) 0.020e-3 (0.586) -0.14e-6 (-0.054) -0.36e-6 (-0.157) 4.76e-05
\n", "\n", "\n", " \n", " \n", " \n", " \n", " 2024-01-31T10:49:33.332534\n", " image/svg+xml\n", " \n", " \n", " Matplotlib v3.7.2, https://matplotlib.org/\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "\n" ], "text/plain": [ "┌─────────────────────────────────────────────────────────────────────────┐\n", "│ Migrad │\n", "├──────────────────────────────────┬──────────────────────────────────────┤\n", "│ FCN = -4.08e+06 │ Nfcn = 120 │\n", "│ EDM = 1.25e-05 (Goal: 0.0002) │ time = 0.9 sec │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ Valid Minimum │ Below EDM threshold (goal x 10) │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ No parameters at limit │ Below call limit │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ Hesse ok │ Covariance accurate │\n", "└──────────────────────────────────┴──────────────────────────────────────┘\n", "┌───┬───────┬───────────┬───────────┬────────────┬────────────┬─────────┬─────────┬───────┐\n", "│ │ Name │ Value │ Hesse Err │ Minos Err- │ Minos Err+ │ Limit- │ Limit+ │ Fixed │\n", "├───┼───────┼───────────┼───────────┼────────────┼────────────┼─────────┼─────────┼───────┤\n", "│ 0 │ s │ 0.995 │ 0.004 │ │ │ 0 │ │ │\n", "│ 1 │ b │ 1.008 │ 0.005 │ │ │ 0 │ │ │\n", "│ 2 │ mu │ 999.3e-3 │ 0.4e-3 │ │ │ 0 │ 2 │ │\n", "│ 3 │ sigma │ 99.21e-3 │ 0.34e-3 │ │ │ 0 │ │ │\n", "│ 4 │ tau │ 1.002 │ 0.007 │ │ │ 0 │ │ │\n", "└───┴───────┴───────────┴───────────┴────────────┴────────────┴─────────┴─────────┴───────┘\n", "┌───────┬───────────────────────────────────────────────────┐\n", "│ │ s b mu sigma tau │\n", "├───────┼───────────────────────────────────────────────────┤\n", "│ s │ 1.32e-05 -0.005e-3 -0.04e-6 0.30e-6 -0.004e-3 │\n", "│ b │ -0.005e-3 2.36e-05 0 -0.46e-6 0.020e-3 │\n", "│ mu │ -0.04e-6 0 1.46e-07 -0.01e-6 -0.14e-6 │\n", "│ sigma │ 0.30e-6 -0.46e-6 -0.01e-6 1.12e-07 -0.36e-6 │\n", "│ tau │ -0.004e-3 0.020e-3 -0.14e-6 -0.36e-6 4.76e-05 │\n", "└───────┴───────────────────────────────────────────────────┘" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "def density(x, s, b, mu, sigma, tau):\n", " xrange = (0, 2)\n", " s1 = s * n * np.diff(norm.cdf(xrange, mu, sigma))\n", " b1 = b * n * np.diff(expon.cdf(xrange, 0, tau))\n", " return s1 + b1, (\n", " s * n * norm.pdf(x, mu, sigma) + \n", " b * n * expon.pdf(x, 0, tau)\n", " )\n", "\n", "m = fit(ExtendedUnbinnedNLL(pts, density))\n", "par_names = [m.params[i].name for i in range(m.npar)]\n", "results = {np.inf: (np.array(m.values), np.array(m.errors), m.fmin.time)}\n", "m" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "xm = np.linspace(np.min(pts), np.max(pts), 1000)\n", "_, ym = density(xm, *m.values)\n", "plt.hist(pts, bins=100, range=(0, 2), label=\"data\")\n", "dx = 2 / 100\n", "plt.plot(xm, ym * dx, label=\"fit\")\n", "plt.legend()\n", "plt.xlabel(\"x\");" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This fit is unbinned, the observed sample is binned here only for visualisation." ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Migrad
FCN = 190.9 (χ²/ndof = 1.0) Nfcn = 110
EDM = 2.17e-06 (Goal: 0.0002)
Valid Minimum Below EDM threshold (goal x 10)
No parameters at limit Below call limit
Hesse ok Covariance accurate
\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Name Value Hesse Error Minos Error- Minos Error+ Limit- Limit+ Fixed
0 s 0.995 0.004 0
1 b 1.008 0.005 0
2 mu 999.3e-3 0.4e-3 0 2
3 sigma 99.22e-3 0.34e-3 0
4 tau 1.002 0.007 0
\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
s b mu sigma tau
s 1.32e-05 -0.005e-3 (-0.284) -0.04e-6 (-0.029) 0.30e-6 (0.248) -0.004e-3 (-0.161)
b -0.005e-3 (-0.284) 2.37e-05 0 -0.47e-6 (-0.286) 0.020e-3 (0.587)
mu -0.04e-6 (-0.029) 0 1.46e-07 -0.01e-6 (-0.046) -0.14e-6 (-0.054)
sigma 0.30e-6 (0.248) -0.47e-6 (-0.286) -0.01e-6 (-0.046) 1.12e-07 -0.37e-6 (-0.159)
tau -0.004e-3 (-0.161) 0.020e-3 (0.587) -0.14e-6 (-0.054) -0.37e-6 (-0.159) 4.77e-05
\n", "\n", "\n", " \n", " \n", " \n", " \n", " 2024-01-31T10:49:34.246748\n", " image/svg+xml\n", " \n", " \n", " Matplotlib v3.7.2, https://matplotlib.org/\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "\n" ], "text/plain": [ "┌─────────────────────────────────────────────────────────────────────────┐\n", "│ Migrad │\n", "├──────────────────────────────────┬──────────────────────────────────────┤\n", "│ FCN = 190.9 (χ²/ndof = 1.0) │ Nfcn = 110 │\n", "│ EDM = 2.17e-06 (Goal: 0.0002) │ │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ Valid Minimum │ Below EDM threshold (goal x 10) │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ No parameters at limit │ Below call limit │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ Hesse ok │ Covariance accurate │\n", "└──────────────────────────────────┴──────────────────────────────────────┘\n", "┌───┬───────┬───────────┬───────────┬────────────┬────────────┬─────────┬─────────┬───────┐\n", "│ │ Name │ Value │ Hesse Err │ Minos Err- │ Minos Err+ │ Limit- │ Limit+ │ Fixed │\n", "├───┼───────┼───────────┼───────────┼────────────┼────────────┼─────────┼─────────┼───────┤\n", "│ 0 │ s │ 0.995 │ 0.004 │ │ │ 0 │ │ │\n", "│ 1 │ b │ 1.008 │ 0.005 │ │ │ 0 │ │ │\n", "│ 2 │ mu │ 999.3e-3 │ 0.4e-3 │ │ │ 0 │ 2 │ │\n", "│ 3 │ sigma │ 99.22e-3 │ 0.34e-3 │ │ │ 0 │ │ │\n", "│ 4 │ tau │ 1.002 │ 0.007 │ │ │ 0 │ │ │\n", "└───┴───────┴───────────┴───────────┴────────────┴────────────┴─────────┴─────────┴───────┘\n", "┌───────┬───────────────────────────────────────────────────┐\n", "│ │ s b mu sigma tau │\n", "├───────┼───────────────────────────────────────────────────┤\n", "│ s │ 1.32e-05 -0.005e-3 -0.04e-6 0.30e-6 -0.004e-3 │\n", "│ b │ -0.005e-3 2.37e-05 0 -0.47e-6 0.020e-3 │\n", "│ mu │ -0.04e-6 0 1.46e-07 -0.01e-6 -0.14e-6 │\n", "│ sigma │ 0.30e-6 -0.47e-6 -0.01e-6 1.12e-07 -0.37e-6 │\n", "│ tau │ -0.004e-3 0.020e-3 -0.14e-6 -0.37e-6 4.77e-05 │\n", "└───────┴───────────────────────────────────────────────────┘" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Migrad
FCN = 87.19 (χ²/ndof = 0.9) Nfcn = 110
EDM = 2.41e-06 (Goal: 0.0002)
Valid Minimum Below EDM threshold (goal x 10)
No parameters at limit Below call limit
Hesse ok Covariance accurate
\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Name Value Hesse Error Minos Error- Minos Error+ Limit- Limit+ Fixed
0 s 0.995 0.004 0
1 b 1.008 0.005 0
2 mu 999.3e-3 0.4e-3 0 2
3 sigma 99.24e-3 0.34e-3 0
4 tau 1.002 0.007 0
\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
s b mu sigma tau
s 1.32e-05 -0.005e-3 (-0.284) -0.04e-6 (-0.029) 0.30e-6 (0.249) -0.004e-3 (-0.161)
b -0.005e-3 (-0.284) 2.37e-05 0 -0.47e-6 (-0.286) 0.020e-3 (0.587)
mu -0.04e-6 (-0.029) 0 1.47e-07 -0.01e-6 (-0.046) -0.14e-6 (-0.054)
sigma 0.30e-6 (0.249) -0.47e-6 (-0.286) -0.01e-6 (-0.046) 1.13e-07 -0.37e-6 (-0.159)
tau -0.004e-3 (-0.161) 0.020e-3 (0.587) -0.14e-6 (-0.054) -0.37e-6 (-0.159) 4.78e-05
\n", "\n", "\n", " \n", " \n", " \n", " \n", " 2024-01-31T10:49:34.512128\n", " image/svg+xml\n", " \n", " \n", " Matplotlib v3.7.2, https://matplotlib.org/\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "\n" ], "text/plain": [ "┌─────────────────────────────────────────────────────────────────────────┐\n", "│ Migrad │\n", "├──────────────────────────────────┬──────────────────────────────────────┤\n", "│ FCN = 87.19 (χ²/ndof = 0.9) │ Nfcn = 110 │\n", "│ EDM = 2.41e-06 (Goal: 0.0002) │ │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ Valid Minimum │ Below EDM threshold (goal x 10) │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ No parameters at limit │ Below call limit │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ Hesse ok │ Covariance accurate │\n", "└──────────────────────────────────┴──────────────────────────────────────┘\n", "┌───┬───────┬───────────┬───────────┬────────────┬────────────┬─────────┬─────────┬───────┐\n", "│ │ Name │ Value │ Hesse Err │ Minos Err- │ Minos Err+ │ Limit- │ Limit+ │ Fixed │\n", "├───┼───────┼───────────┼───────────┼────────────┼────────────┼─────────┼─────────┼───────┤\n", "│ 0 │ s │ 0.995 │ 0.004 │ │ │ 0 │ │ │\n", "│ 1 │ b │ 1.008 │ 0.005 │ │ │ 0 │ │ │\n", "│ 2 │ mu │ 999.3e-3 │ 0.4e-3 │ │ │ 0 │ 2 │ │\n", "│ 3 │ sigma │ 99.24e-3 │ 0.34e-3 │ │ │ 0 │ │ │\n", "│ 4 │ tau │ 1.002 │ 0.007 │ │ │ 0 │ │ │\n", "└───┴───────┴───────────┴───────────┴────────────┴────────────┴─────────┴─────────┴───────┘\n", "┌───────┬───────────────────────────────────────────────────┐\n", "│ │ s b mu sigma tau │\n", "├───────┼───────────────────────────────────────────────────┤\n", "│ s │ 1.32e-05 -0.005e-3 -0.04e-6 0.30e-6 -0.004e-3 │\n", "│ b │ -0.005e-3 2.37e-05 0 -0.47e-6 0.020e-3 │\n", "│ mu │ -0.04e-6 0 1.47e-07 -0.01e-6 -0.14e-6 │\n", "│ sigma │ 0.30e-6 -0.47e-6 -0.01e-6 1.13e-07 -0.37e-6 │\n", "│ tau │ -0.004e-3 0.020e-3 -0.14e-6 -0.37e-6 4.78e-05 │\n", "└───────┴───────────────────────────────────────────────────┘" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Migrad
FCN = 36.34 (χ²/ndof = 0.8) Nfcn = 110
EDM = 1.7e-06 (Goal: 0.0002)
Valid Minimum Below EDM threshold (goal x 10)
No parameters at limit Below call limit
Hesse ok Covariance accurate
\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Name Value Hesse Error Minos Error- Minos Error+ Limit- Limit+ Fixed
0 s 0.995 0.004 0
1 b 1.008 0.005 0
2 mu 999.3e-3 0.4e-3 0 2
3 sigma 99.20e-3 0.34e-3 0
4 tau 1.002 0.007 0
\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
s b mu sigma tau
s 1.32e-05 -0.005e-3 (-0.285) -0.04e-6 (-0.029) 0.31e-6 (0.250) -0.004e-3 (-0.162)
b -0.005e-3 (-0.285) 2.38e-05 0 -0.48e-6 (-0.287) 0.020e-3 (0.588)
mu -0.04e-6 (-0.029) 0 1.48e-07 -0.01e-6 (-0.046) -0.15e-6 (-0.055)
sigma 0.31e-6 (0.250) -0.48e-6 (-0.287) -0.01e-6 (-0.046) 1.16e-07 -0.38e-6 (-0.160)
tau -0.004e-3 (-0.162) 0.020e-3 (0.588) -0.15e-6 (-0.055) -0.38e-6 (-0.160) 4.79e-05
\n", "\n", "\n", " \n", " \n", " \n", " \n", " 2024-01-31T10:49:34.740552\n", " image/svg+xml\n", " \n", " \n", " Matplotlib v3.7.2, https://matplotlib.org/\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "\n" ], "text/plain": [ "┌─────────────────────────────────────────────────────────────────────────┐\n", "│ Migrad │\n", "├──────────────────────────────────┬──────────────────────────────────────┤\n", "│ FCN = 36.34 (χ²/ndof = 0.8) │ Nfcn = 110 │\n", "│ EDM = 1.7e-06 (Goal: 0.0002) │ │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ Valid Minimum │ Below EDM threshold (goal x 10) │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ No parameters at limit │ Below call limit │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ Hesse ok │ Covariance accurate │\n", "└──────────────────────────────────┴──────────────────────────────────────┘\n", "┌───┬───────┬───────────┬───────────┬────────────┬────────────┬─────────┬─────────┬───────┐\n", "│ │ Name │ Value │ Hesse Err │ Minos Err- │ Minos Err+ │ Limit- │ Limit+ │ Fixed │\n", "├───┼───────┼───────────┼───────────┼────────────┼────────────┼─────────┼─────────┼───────┤\n", "│ 0 │ s │ 0.995 │ 0.004 │ │ │ 0 │ │ │\n", "│ 1 │ b │ 1.008 │ 0.005 │ │ │ 0 │ │ │\n", "│ 2 │ mu │ 999.3e-3 │ 0.4e-3 │ │ │ 0 │ 2 │ │\n", "│ 3 │ sigma │ 99.20e-3 │ 0.34e-3 │ │ │ 0 │ │ │\n", "│ 4 │ tau │ 1.002 │ 0.007 │ │ │ 0 │ │ │\n", "└───┴───────┴───────────┴───────────┴────────────┴────────────┴─────────┴─────────┴───────┘\n", "┌───────┬───────────────────────────────────────────────────┐\n", "│ │ s b mu sigma tau │\n", "├───────┼───────────────────────────────────────────────────┤\n", "│ s │ 1.32e-05 -0.005e-3 -0.04e-6 0.31e-6 -0.004e-3 │\n", "│ b │ -0.005e-3 2.38e-05 0 -0.48e-6 0.020e-3 │\n", "│ mu │ -0.04e-6 0 1.48e-07 -0.01e-6 -0.15e-6 │\n", "│ sigma │ 0.31e-6 -0.48e-6 -0.01e-6 1.16e-07 -0.38e-6 │\n", "│ tau │ -0.004e-3 0.020e-3 -0.15e-6 -0.38e-6 4.79e-05 │\n", "└───────┴───────────────────────────────────────────────────┘" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Migrad
FCN = 7.555 (χ²/ndof = 0.5) Nfcn = 102
EDM = 5.22e-05 (Goal: 0.0002)
Valid Minimum Below EDM threshold (goal x 10)
No parameters at limit Below call limit
Hesse ok Covariance accurate
\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Name Value Hesse Error Minos Error- Minos Error+ Limit- Limit+ Fixed
0 s 0.995 0.004 0
1 b 1.008 0.005 0
2 mu 999.2e-3 0.4e-3 0 2
3 sigma 99.5e-3 0.4e-3 0
4 tau 1.002 0.007 0
\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
s b mu sigma tau
s 1.34e-05 -0.005e-3 (-0.297) -0.04e-6 (-0.030) 0.35e-6 (0.260) -0.004e-3 (-0.169)
b -0.005e-3 (-0.297) 2.42e-05 -0 -0.54e-6 (-0.298) 0.020e-3 (0.591)
mu -0.04e-6 (-0.030) -0 1.61e-07 -0.01e-6 (-0.048) -0.16e-6 (-0.059)
sigma 0.35e-6 (0.260) -0.54e-6 (-0.298) -0.01e-6 (-0.048) 1.37e-07 -0.43e-6 (-0.167)
tau -0.004e-3 (-0.169) 0.020e-3 (0.591) -0.16e-6 (-0.059) -0.43e-6 (-0.167) 4.83e-05
\n", "\n", "\n", " \n", " \n", " \n", " \n", " 2024-01-31T10:49:34.939467\n", " image/svg+xml\n", " \n", " \n", " Matplotlib v3.7.2, https://matplotlib.org/\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "\n" ], "text/plain": [ "┌─────────────────────────────────────────────────────────────────────────┐\n", "│ Migrad │\n", "├──────────────────────────────────┬──────────────────────────────────────┤\n", "│ FCN = 7.555 (χ²/ndof = 0.5) │ Nfcn = 102 │\n", "│ EDM = 5.22e-05 (Goal: 0.0002) │ │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ Valid Minimum │ Below EDM threshold (goal x 10) │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ No parameters at limit │ Below call limit │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ Hesse ok │ Covariance accurate │\n", "└──────────────────────────────────┴──────────────────────────────────────┘\n", "┌───┬───────┬───────────┬───────────┬────────────┬────────────┬─────────┬─────────┬───────┐\n", "│ │ Name │ Value │ Hesse Err │ Minos Err- │ Minos Err+ │ Limit- │ Limit+ │ Fixed │\n", "├───┼───────┼───────────┼───────────┼────────────┼────────────┼─────────┼─────────┼───────┤\n", "│ 0 │ s │ 0.995 │ 0.004 │ │ │ 0 │ │ │\n", "│ 1 │ b │ 1.008 │ 0.005 │ │ │ 0 │ │ │\n", "│ 2 │ mu │ 999.2e-3 │ 0.4e-3 │ │ │ 0 │ 2 │ │\n", "│ 3 │ sigma │ 99.5e-3 │ 0.4e-3 │ │ │ 0 │ │ │\n", "│ 4 │ tau │ 1.002 │ 0.007 │ │ │ 0 │ │ │\n", "└───┴───────┴───────────┴───────────┴────────────┴────────────┴─────────┴─────────┴───────┘\n", "┌───────┬───────────────────────────────────────────────────┐\n", "│ │ s b mu sigma tau │\n", "├───────┼───────────────────────────────────────────────────┤\n", "│ s │ 1.34e-05 -0.005e-3 -0.04e-6 0.35e-6 -0.004e-3 │\n", "│ b │ -0.005e-3 2.42e-05 -0 -0.54e-6 0.020e-3 │\n", "│ mu │ -0.04e-6 -0 1.61e-07 -0.01e-6 -0.16e-6 │\n", "│ sigma │ 0.35e-6 -0.54e-6 -0.01e-6 1.37e-07 -0.43e-6 │\n", "│ tau │ -0.004e-3 0.020e-3 -0.16e-6 -0.43e-6 4.83e-05 │\n", "└───────┴───────────────────────────────────────────────────┘" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Migrad
FCN = 2.084 (χ²/ndof = 0.4) Nfcn = 114
EDM = 4.65e-08 (Goal: 0.0002)
Valid Minimum Below EDM threshold (goal x 10)
No parameters at limit Below call limit
Hesse ok Covariance accurate
\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Name Value Hesse Error Minos Error- Minos Error+ Limit- Limit+ Fixed
0 s 0.995 0.004 0
1 b 1.007 0.005 0
2 mu 999.4e-3 0.5e-3 0 2
3 sigma 99.8e-3 0.7e-3 0
4 tau 1.003 0.007 0
\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
s b mu sigma tau
s 1.54e-05 -0.008e-3 (-0.397) -0.07e-6 (-0.042) 1.1e-6 (0.413) -0.007e-3 (-0.238)
b -0.008e-3 (-0.397) 2.9e-05 0.02e-6 (0.009) -1.8e-6 (-0.464) 0.024e-3 (0.624)
mu -0.07e-6 (-0.042) 0.02e-6 (0.009) 2.06e-07 -0.02e-6 (-0.071) -0.21e-6 (-0.063)
sigma 1.1e-6 (0.413) -1.8e-6 (-0.464) -0.02e-6 (-0.071) 4.95e-07 -1.4e-6 (-0.275)
tau -0.007e-3 (-0.238) 0.024e-3 (0.624) -0.21e-6 (-0.063) -1.4e-6 (-0.275) 5.22e-05
\n", "\n", "\n", " \n", " \n", " \n", " \n", " 2024-01-31T10:49:35.124271\n", " image/svg+xml\n", " \n", " \n", " Matplotlib v3.7.2, https://matplotlib.org/\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "\n" ], "text/plain": [ "┌─────────────────────────────────────────────────────────────────────────┐\n", "│ Migrad │\n", "├──────────────────────────────────┬──────────────────────────────────────┤\n", "│ FCN = 2.084 (χ²/ndof = 0.4) │ Nfcn = 114 │\n", "│ EDM = 4.65e-08 (Goal: 0.0002) │ │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ Valid Minimum │ Below EDM threshold (goal x 10) │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ No parameters at limit │ Below call limit │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ Hesse ok │ Covariance accurate │\n", "└──────────────────────────────────┴──────────────────────────────────────┘\n", "┌───┬───────┬───────────┬───────────┬────────────┬────────────┬─────────┬─────────┬───────┐\n", "│ │ Name │ Value │ Hesse Err │ Minos Err- │ Minos Err+ │ Limit- │ Limit+ │ Fixed │\n", "├───┼───────┼───────────┼───────────┼────────────┼────────────┼─────────┼─────────┼───────┤\n", "│ 0 │ s │ 0.995 │ 0.004 │ │ │ 0 │ │ │\n", "│ 1 │ b │ 1.007 │ 0.005 │ │ │ 0 │ │ │\n", "│ 2 │ mu │ 999.4e-3 │ 0.5e-3 │ │ │ 0 │ 2 │ │\n", "│ 3 │ sigma │ 99.8e-3 │ 0.7e-3 │ │ │ 0 │ │ │\n", "│ 4 │ tau │ 1.003 │ 0.007 │ │ │ 0 │ │ │\n", "└───┴───────┴───────────┴───────────┴────────────┴────────────┴─────────┴─────────┴───────┘\n", "┌───────┬───────────────────────────────────────────────────┐\n", "│ │ s b mu sigma tau │\n", "├───────┼───────────────────────────────────────────────────┤\n", "│ s │ 1.54e-05 -0.008e-3 -0.07e-6 1.1e-6 -0.007e-3 │\n", "│ b │ -0.008e-3 2.9e-05 0.02e-6 -1.8e-6 0.024e-3 │\n", "│ mu │ -0.07e-6 0.02e-6 2.06e-07 -0.02e-6 -0.21e-6 │\n", "│ sigma │ 1.1e-6 -1.8e-6 -0.02e-6 4.95e-07 -1.4e-6 │\n", "│ tau │ -0.007e-3 0.024e-3 -0.21e-6 -1.4e-6 5.22e-05 │\n", "└───────┴───────────────────────────────────────────────────┘" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Migrad
FCN = 2.677e-06 Nfcn = 112
EDM = 2.68e-06 (Goal: 0.0002)
Valid Minimum Below EDM threshold (goal x 10)
No parameters at limit Below call limit
Hesse ok Covariance accurate
\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
Name Value Hesse Error Minos Error- Minos Error+ Limit- Limit+ Fixed
0 s 0.996 0.005 0
1 b 1.006 0.007 0
2 mu 0.9998 0.0019 0 2
3 sigma 0.1001 0.0012 0
4 tau 1.002 0.008 0
\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
s b mu sigma tau
s 2.31e-05 -0.020e-3 (-0.599) -2e-6 (-0.173) 3.8e-6 (0.663) -0.014e-3 (-0.350)
b -0.020e-3 (-0.599) 4.64e-05 1e-6 (0.041) -5.5e-6 (-0.684) 0.04e-3 (0.673)
mu -2e-6 (-0.173) 1e-6 (0.041) 3.72e-06 -0.7e-6 (-0.291) -4e-6 (-0.259)
sigma 3.8e-6 (0.663) -5.5e-6 (-0.684) -0.7e-6 (-0.291) 1.4e-06 -3.7e-6 (-0.369)
tau -0.014e-3 (-0.350) 0.04e-3 (0.673) -4e-6 (-0.259) -3.7e-6 (-0.369) 7.03e-05
\n", "\n", "\n", " \n", " \n", " \n", " \n", " 2024-01-31T10:49:35.369416\n", " image/svg+xml\n", " \n", " \n", " Matplotlib v3.7.2, https://matplotlib.org/\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "\n" ], "text/plain": [ "┌─────────────────────────────────────────────────────────────────────────┐\n", "│ Migrad │\n", "├──────────────────────────────────┬──────────────────────────────────────┤\n", "│ FCN = 2.677e-06 │ Nfcn = 112 │\n", "│ EDM = 2.68e-06 (Goal: 0.0002) │ │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ Valid Minimum │ Below EDM threshold (goal x 10) │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ No parameters at limit │ Below call limit │\n", "├──────────────────────────────────┼──────────────────────────────────────┤\n", "│ Hesse ok │ Covariance accurate │\n", "└──────────────────────────────────┴──────────────────────────────────────┘\n", "┌───┬───────┬───────────┬───────────┬────────────┬────────────┬─────────┬─────────┬───────┐\n", "│ │ Name │ Value │ Hesse Err │ Minos Err- │ Minos Err+ │ Limit- │ Limit+ │ Fixed │\n", "├───┼───────┼───────────┼───────────┼────────────┼────────────┼─────────┼─────────┼───────┤\n", "│ 0 │ s │ 0.996 │ 0.005 │ │ │ 0 │ │ │\n", "│ 1 │ b │ 1.006 │ 0.007 │ │ │ 0 │ │ │\n", "│ 2 │ mu │ 0.9998 │ 0.0019 │ │ │ 0 │ 2 │ │\n", "│ 3 │ sigma │ 0.1001 │ 0.0012 │ │ │ 0 │ │ │\n", "│ 4 │ tau │ 1.002 │ 0.008 │ │ │ 0 │ │ │\n", "└───┴───────┴───────────┴───────────┴────────────┴────────────┴─────────┴─────────┴───────┘\n", "┌───────┬───────────────────────────────────────────────────┐\n", "│ │ s b mu sigma tau │\n", "├───────┼───────────────────────────────────────────────────┤\n", "│ s │ 2.31e-05 -0.020e-3 -2e-6 3.8e-6 -0.014e-3 │\n", "│ b │ -0.020e-3 4.64e-05 1e-6 -5.5e-6 0.04e-3 │\n", "│ mu │ -2e-6 1e-6 3.72e-06 -0.7e-6 -4e-6 │\n", "│ sigma │ 3.8e-6 -5.5e-6 -0.7e-6 1.4e-06 -3.7e-6 │\n", "│ tau │ -0.014e-3 0.04e-3 -4e-6 -3.7e-6 7.03e-05 │\n", "└───────┴───────────────────────────────────────────────────┘" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "def integral(xe, s, b, mu, sigma, tau):\n", " return s * n * norm.cdf(xe, mu, sigma) + b * n * expon.cdf(xe, 0, tau)\n", "\n", "fig, ax = plt.subplots(3, 2, figsize=(10, 8), sharex=True, constrained_layout=True)\n", "for axi, bins in zip(ax.flat, (200, 100, 50, 20, 10, 5)):\n", " w, xe = np.histogram(pts, bins=bins, range=(0, 2))\n", " c = ExtendedBinnedNLL(w, xe, integral)\n", " m = fit(c)\n", " display(m.fmin)\n", " axi.stairs(w, xe, fill=True, label=\"data\")\n", " axi.stairs(np.diff(integral(xe, *m.values)), xe, label=\"fit\")\n", " axi.legend(title=f\"bins = {len(w)}\")\n", " results[bins] = (np.array(m.values), np.array(m.errors), m.fmin.time)\n", "fig.supxlabel(\"x\");" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "npar = len(results[np.inf][0])\n", "\n", "fig, ax = plt.subplots(npar, 2, sharex=True, figsize=(14, 20))\n", "for j, (k, (v, e, _)) in enumerate(results.items()):\n", " for i, (vi, ei) in enumerate(zip(v, e)):\n", " c = f\"C{i}\"\n", " ax[i, 0].errorbar(j, vi, ei, color=c, fmt=\"o\")\n", " ax[i, 0].set_ylabel(par_names[i])\n", " einf = results[np.inf][1][i]\n", " ax[i, 1].plot(j, ei /einf, \"o\", color=c)\n", "for i in range(npar):\n", " ax[i, 1].set_ylim(0.95, 1.2)\n", " ax[i, 1].axhline(1.05, ls=\"--\", color=\"0.5\")\n", "plt.xticks(np.arange(7), [f\"{x}\" for x in results.keys()]);" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Shown on the left is the fitted value and its uncertainty estimate. Shown of the right is the relative size of the error bar of the binned fit compared to the unbinned fit." ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plt.figure()\n", "x = np.arange(7)\n", "y = [v[2] for v in results.values()]\n", "plt.plot(x, y, \"o\")\n", "for xi, yi in zip(x[1:], y[1:]):\n", " plt.text(xi, yi * 1.2, f\"{y[0]/yi:.0f}x\", ha=\"center\")\n", "plt.xticks(x, [f\"{x}\" for x in results.keys()])\n", "plt.ylabel(\"time / sec\")\n", "plt.semilogy();" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We now demonstrate that the binned fits and the unbinned fit are unbiased. We repeat the fit many times with independent random samples, the mean of the results minus the truth is the bias. In each iteration, the binned fits use the same data that the unbinned fit uses." ] }, { "cell_type": "code", "execution_count": 20, "metadata": {}, "outputs": [], "source": [ "@joblib.delayed\n", "def run(seed):\n", " rng = np.random.default_rng(seed)\n", " s = rng.normal(truth[2], truth[3], size=int(n * truth[0]))\n", " b = rng.exponential(truth[4], size=int(n * truth[1]))\n", " pts = np.append(s, b)\n", " pts = pts[(pts > 0) & (pts < 2)]\n", "\n", " if bins == np.inf:\n", " m = fit(ExtendedUnbinnedNLL(pts, density))\n", " assert m.valid\n", " else:\n", " w, xe = np.histogram(pts, bins=bins, range=(0, 2))\n", " m = fit(ExtendedBinnedNLL(w, xe, integral))\n", " assert m.valid\n", " return np.array(m.values)\n", "\n", "results = {}\n", "for bins in (np.inf, 200, 100, 50, 20, 10, 5):\n", " results[bins] = joblib.Parallel(-1)(run(seed) for seed in range(100))" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAi8AAAGzCAYAAADnmPfhAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAAAn3klEQVR4nO3de1xVdaL///fmLip4gQCVEUwmJe8XHNQGmviK5inMGacsRc30pOFPh5rSjnmpztBkXqjxMtWjcB7azXmU1rE0MzVHEfLCmRwvmaE4CYiZoJIgsH9/dNi5hY0obDYfeD0fj/14uNb+rMVHV+z9aq21wWK1Wq0CAAAwhJurJwAAAHAjiBcAAGAU4gUAABiFeAEAAEYhXgAAgFGIFwAAYBTiBQAAGIV4AQAARiFeAACAUYgXAHWyYMECWSwWnT17tsZxYWFhmjhxYsNMCkCTRrwAaJKKi4u1fPlyDRs2TCEhIWrdurX69u2rlStXqry8vMr4iooKvfjiiwoPD5ePj4969eqlt99+2wUzB3A9Hq6eAIDm4ejRo3Jza7j/X/r22281Y8YM3XXXXUpOTpafn582b96s6dOna8+ePVq9erXd+P/6r//SCy+8oClTpmjgwIHasGGDHnzwQVksFj3wwAMNNm8A12fhFzMCqIsFCxZo4cKFKigoUEBAgKunY3P27Fnl5+fr9ttvt1v/8MMP680339SxY8fUtWtXSdJ3332n8PBwTZ06VX/5y18kSVarVTExMcrOztaJEyfk7u7e4H8HANXjshGAenH27Fn9/ve/l5+fn9q3b6+ZM2fq8uXLtuevveclLS1NFotFu3btUnJysgIDA9WyZUvdd999KigoqPN8AgICqoSLJN13332SpMOHD9vWbdiwQVeuXNH06dNt6ywWi6ZNm6Z///vfSk9Pr/N8ANQfLhsBqBe///3vFRYWppSUFO3Zs0cvv/yyfvjhB/3tb3+rcbsZM2aobdu2mj9/vk6cOKFly5YpKSlJ7777rm3MxYsX7ULIEU9PT/n7+9c4Ji8vT5LszhIdOHBALVu2VPfu3e3GRkVF2Z4fOnTodb8+gIZBvACoF+Hh4dqwYYMk6bHHHpOfn59WrFihJ554Qr169XK4Xfv27fXpp5/KYrFI+unG2ZdfflmFhYW2EElKSqpyj0p1YmJitH37dofPl5aWatmyZQoPD9fAgQNt63NzcxUUFGSbQ6WQkBBJ0unTp6/7tQE0HOIFQL147LHH7JZnzJihFStW6OOPP64xXqZOnWoXDXfccYeWLl2qkydP2rZ78sknNW7cuOvOoW3btjU+n5SUpEOHDmnjxo3y8Pj55e/HH3+Ut7d3lfE+Pj625wE0HsQLgHoRERFht3zrrbfKzc1NJ06cqHG7X/ziF3bLlQHyww8/2NZFRkYqMjKyTvNbtGiRXnvtNT333HO6++677Z5r0aKFSkpKqmxTeamqRYsWdfraAOoX8QLAKa69BOOIo0/xXP1ByMLCwlqd/fDy8lK7du2qrE9LS9NTTz2lRx99VHPnzq3yfEhIiLZt2yar1Wo379zcXElShw4drvu1ATQcPm0EoF4cO3bMbvmbb75RRUWFwsLC6rzvmTNnKiQk5LqP0aNHV9l2w4YNeuSRRzR69GgtX7682v336dNHxcXFdp9AkqSMjAzb8wAaD868AKgXlT/NttIrr7wiSRoxYkSd932z97x88cUXeuCBB/TrX/9aa9eudfhD8hISEvSHP/xBK1assPs5L6tWrVLHjh01ePDgOv8dANQf4gVAvcjOzta9996r4cOHKz09XWvWrNGDDz6o3r1713nfN3PPy8mTJ3XvvffKYrHod7/7ndatW2f3fK9evWw3BHfq1EmzZs3SokWLdOXKFQ0cOFDr16/Xzp07tXbtWn5AHdDIEC8A6sW7776refPmafbs2fLw8FBSUpIWLVrksvlkZ2ersLBQUtVPQknS/Pnz7T4F9cILL6ht27b661//qrS0NEVERNgCDEDjwq8HAAAARuGGXQAAYBTiBQAAGIV4AQAARiFeAACAUYgXAABgFOIFAAAYpcn9nJeKigqdPn1arVu3rvXvVgEAAK5ltVp14cIFdejQweFPw67U5OLl9OnTCg0NdfU0AADATTh16pQ6depU45gmFy+tW7eW9NNf3s/Pz8WzAQAAtVFUVKTQ0FDb+3hNmly8VF4q8vPzI14AADBMbW754IZdAABgFOIFAAAYhXgBAABGIV4AAIBRiBcAAGAU4gUAABiFeAEAAEYhXgAAgFGIFwAAYBTiBQAAGIV4AQAARiFeAACAUYgXAABgFOIFzUZxaZnCZm9U2OyNKi4tc/V0AAA3iXgBAABGIV4AAIBRiBcAAGAU4gUAABiFeAEAAEYhXgAAgFGIFwAAYBTiBQAAGIV4AQAARiFeAACAUYgXAABgFOIFAAAYhXgBAABGIV4AAIBRnBovX3zxhe655x516NBBFotF69evv+4227dvV79+/eTt7a2uXbsqLS3NmVMEAACGcWq8XLp0Sb1799by5ctrNT47O1sjR47UnXfeqaysLM2aNUuPPPKINm/e7MxpAgAAg3g4c+cjRozQiBEjaj1+1apVCg8P1+LFiyVJ3bt31z/+8Q8tXbpU8fHx1W5TUlKikpIS23JRUVHdJg0AABq1RnXPS3p6uuLi4uzWxcfHKz093eE2KSkp8vf3tz1CQ0OdPU0AAJql4tIyhc3eqLDZG1VcWuayeTSqeMnLy1NQUJDduqCgIBUVFenHH3+sdps5c+aosLDQ9jh16lRDTBUAALiIUy8bNQRvb295e3u7ehoAAKCBNKozL8HBwcrPz7dbl5+fLz8/P7Vo0cJFswIAAI1Jo4qX6Ohobd261W7dli1bFB0d7aIZAQDqS2O5XwLmc2q8XLx4UVlZWcrKypL000ehs7KylJOTI+mn+1USExNt4x999FF9++23evLJJ3XkyBGtWLFC7733nv7whz84c5oADMAbH4BKTo2XvXv3qm/fvurbt68kKTk5WX379tW8efMkSbm5ubaQkaTw8HBt3LhRW7ZsUe/evbV48WK9/vrrDj8mDQAAmh+n3rAbGxsrq9Xq8PnqfnpubGysDhw44MRZAQAAkzWqe14AAACuh3gBAABGIV4AAIBRiBcAAGAU4gUAABiFeAEAAEYhXgAAgFGIFwAAYBTiBQAAGIV4AQAARiFeAACAUYgXAABgFOIFAAAYhXgBAABGIV4AAIBRiJdaKi4tU9jsjQqbvVHFpWWung4AAM0W8QIAAIxCvAAAAKMQLwAAwCjECwAAMArxAgAAjEK8AAAAoxAvAADAKMQLAAAwCvECAACMQrwAAACjEC8AAMAoxAsAADAK8QIAAIxCvAAAAKMQLwAAwCjECwAAMArxAgAAjEK8AAAAoxAvAADAKMQLAAAwCvECAACMQrwAAACjEC8AAMAoxAsAADAK8QIAAIxCvAAAAKMQLwAAwCjECwAAMArxAgAAjEK8AAAAoxAvAADAKMQLAAAwCvECAACMQrwAAACjEC8AAMAoxAsAADAK8QIAAIxCvAAAAKMQLwAAwCjECwAAMArxAgAAjEK8AAAAoxAvAADAKMQLAAAwCvECAACMQrwAAACjNEi8LF++XGFhYfLx8dGgQYOUmZnpcGxaWposFovdw8fHpyGmCQAADOD0eHn33XeVnJys+fPna//+/erdu7fi4+N15swZh9v4+fkpNzfX9jh58qSzpwkAAAzh9HhZsmSJpkyZokmTJikyMlKrVq2Sr6+v3njjDYfbWCwWBQcH2x5BQUHOniYAADCEU+OltLRU+/btU1xc3M9f0M1NcXFxSk9Pd7jdxYsX1blzZ4WGhiohIUH/+te/HI4tKSlRUVGR3QMAADRdTo2Xs2fPqry8vMqZk6CgIOXl5VW7zW233aY33nhDGzZs0Jo1a1RRUaHBgwfr3//+d7XjU1JS5O/vb3uEhobW+98DAAA0Ho3u00bR0dFKTExUnz59FBMTo/fff1+BgYH661//Wu34OXPmqLCw0PY4depUA88YAAA0JA9n7jwgIEDu7u7Kz8+3W5+fn6/g4OBa7cPT01N9+/bVN998U+3z3t7e8vb2rvNcAQCAGZx65sXLy0v9+/fX1q1bbesqKiq0detWRUdH12of5eXl+uqrrxQSEuKsaQIAAIM49cyLJCUnJ2vChAkaMGCAoqKitGzZMl26dEmTJk2SJCUmJqpjx45KSUmRJD377LP61a9+pa5du+r8+fNatGiRTp48qUceecTZUwUAAAZwerzcf//9Kigo0Lx585SXl6c+ffpo06ZNtpt4c3Jy5Ob28wmgH374QVOmTFFeXp7atm2r/v37a/fu3YqMjHT2VAEAgAGcHi+SlJSUpKSkpGqf2759u93y0qVLtXTp0gaYFQAAMFGj+7QRAABATYgXAABgFOIFAAAYhXgBAABGIV4AAIBRiBcAAGAU4gUAABiFeAEAAEYhXgAAgFGIFwAAYBTiBQAAGIV4AQAARiFeAACAUYgXAABgFOIFAAAYhXgBAABGIV4AAIBRiBcAAGAU4gUAABiFeAEAAEYhXgAAgFGIFwAAYBTiBQAAGIV4AQAARiFeAACAUYgXAABgFOIFAAAYhXgBAABGIV4AAIBRiBcAAGAU4gUAABiFeAEAAEYhXgAAgFGIFwAAYBTiBQAAGIV4AQAARiFeAACAUYgXAABgFOIFAAAYhXgBAABGIV4AAIBRiBcAAGAU4gUAABiFeAEAAEYhXtBslFdYbX/OzD5ntwwAMAfxgmZh08FcxS3ZYVue+OaXGvrnz7XpYK4LZwUAuBnEC5q8TQdzNW3NfuUXlditzyu8rGlr9hMwAGAY4gVNWnmFVQs/OqTqLhBVrlv40SEuIQGAQYgXNGmZ2eeUW3jZ4fNWSbmFl5WZfa7hJgUAqBPiBU3amQuOw+VmxgEAXI94QZN2S2ufeh0HAHA94gVNWlR4O4X4+8ji4HmLpBB/H0WFt2vIaQEA6oB4QZPm7mbR/HsiJalKwFQuz78nUu5ujvIGANDYEC9o8ob3CNHKcf10i5+33fpgfx+tHNdPw3uEuGhmAICb4eHqCQANYXiPEA3pGqCeCz6VJKVNGqg7IgI54wIABuLMC5qNq0MlKrwd4QIAhiJeAACAUYgXAABgFOIFAAAYhXgBAABGIV4AAIBRiBcAAFAr5RVW258zs8/ZLTck4gUAAFzXpoO5iluyw7Y88c0vNfTPn2vTwdwGn0uDxMvy5csVFhYmHx8fDRo0SJmZmTWOX7dunbp16yYfHx/17NlTH3/8cUNMEwAAVGPTwVxNW7Nf+UUlduvzCi9r2pr9DR4wTo+Xd999V8nJyZo/f77279+v3r17Kz4+XmfOnKl2/O7duzV27FhNnjxZBw4c0KhRozRq1CgdPHjQ2VMFAADXKK+wauFHh1TdBaLKdQs/OtSgl5Cc/usBlixZoilTpmjSpEmSpFWrVmnjxo164403NHv27CrjU1NTNXz4cP3xj3+UJD333HPasmWL/vKXv2jVqlW1/rqlpaUqLS2tst7NzU0eHh524xyxWCzy9PS0LXuo3LaNhypqHHvlyhVZrdUfSGeNlSQvL6+bGltWVqaKiop6Gevp6SmLxeLUseXl5SovL7+hsaWlZdUeQw8PD7m5udVqvw0xtqKiQmVlZQ7Huru7y93dvdGMtVqtunLlSr2Mvfr789qx1x4/Lzc5HFvTfiu3r4+x135/3sjY5vga8ePlK3JTuSrkpszsc7ojIlDWivJG8xrhSHN/jcjIPqfcwssO92GVlFt4WbuP5WtQeLsq+72R14jacmq8lJaWat++fZozZ45tnZubm+Li4pSenl7tNunp6UpOTrZbFx8fr/Xr11c7vqSkRCUlP5/GKioqkiQtXrxYPj4+VcZHRETowQcftC2/9NJLDv9RO3furIkTJ9qWx/h8JR9LmVIXH6gytkOHDpoyZYptefny5SosLKx2v4GBgZo+fbpt+bXXXlNBQUG1Y/39/TVr1izbclpamk6fPl3tWF9fX1v0SdLatWt18uTJasd6enrq6aefti2/9957OnbsWLVjJWn+/Pm2P3/wwQc6dOiQw7Fz5syxvZD9z//8j/73f//X4dgnnnhCLVu2lCRt3rxZe/fudTh25syZatOmjSRp69atDv8bkqRp06bplltukSTt3LlTO3b8dJ12fIufnr/6GD7yyCPq2LGjJGnPnj367LPPHO53woQJCgsLkyTt27dPn3zyicOxY8eO1S9/+UtJ0ldffaUNGzY4HPu73/1Ot99+uyTp8OHD+vvf/+5wbEJCgvr06SNJ+uabb/T22287HDtixAhFRUVJknJycrR69WqHY+Pi4jRkyBBJUm5url5//XWHY2NiYhQbGytJKigo0MqVKx2OjY6O1rBhwyRJhYWFSk1NdTh2wIABGjlypCSpuLhYL730ku25CqsU59VaP1o9NfvFVRrWt4tG3zdK0k9vwikpKQ73GxkZqTFjxtiWaxpbl9eI1NRUFRcXVzu2ub9GnChvo4zSX6hCP702THzzS4X4+2h40EXpVJbD/Tb0a0R1mvtrxLdl7SR1cbiPSq+vWafPPc5JuvnXiNpyarycPXtW5eXlCgoKslsfFBSkI0eOVLtNXl5etePz8vKqHZ+SkqKFCxfWz4QBNEqVb3zF+vn/7vccsMo3IpffCm6AE+VttK301irr8wov681Cd93p1UZh7ucbfmKolRYWx2dNbmZcfbBYazpnWEenT59Wx44dtXv3bkVHR9vWP/nkk9qxY4cyMjKqbOPl5aXVq1dr7NixtnUrVqzQwoULlZ+fX2V8dWdeQkNDVVBQID8/vyrjb/aUcHFpmXrN++nG4b1z4+Tr5eFwrNQ8TwlXaqyXjYpLyzTg+Z/+j+nqY9jcTwnXdayzLxttPpSv/++df1a53l75azVXjuun+NuDuWxUi7FSw79GlFdYdeeSncq75kZP2/wlBft56/PkO6r9ZalcNqo6tqG/7yuPYX5RSbX3vVR3DG/mNaKoqEj+/v4qLCys9v37ak498xIQECB3d/cq0ZGfn6/g4OBqtwkODr6h8d7e3vL29q6y3svLy+6byZHajKlUJver9l3zP93VLybX0xjGXv1ibcLYG7lGWjm2TG7XPYY3s9/6Huvm5lbr/y4bw1iLxeK0se4envrTJ187vFHQop9uFPx/kcE39L3cGMY2hu/7hniNSD/+vcNwkf7vfomiEmV9d1HRt7avtzk0hu/lpvQaseDe2zVtzX5ZJLvvx8rcnH/v7WrhU/W9WLqx7/vacuqnjby8vNS/f39t3brVtq6iokJbt261OxNztejoaLvxkrRlyxaH4wE0XZm1vFEwM/tcw00KN+TMBcfH72bGwTWG9wjRynH9dIuffaAE+/to5bh+DX751umfNkpOTtaECRM0YMAARUVFadmyZbp06ZLt00eJiYnq2LGj7Qa6mTNnKiYmRosXL9bIkSP1zjvvaO/evXr11VedPVUAjQxvfOa7pXXVD07UZRxcZ3iPEA3pGqCeCz6VJKVNGqg7IgKrvdznbE6Pl/vvv18FBQWaN2+e8vLy1KdPH23atMl2U25OTo7tOp4kDR48WG+99Zbmzp2rp59+WhEREVq/fr169Ojh7KkCaGR44zNfVHg7hfj7KK/wsuP7Jfx9FPV/H7FF43Z1qESFt3NJuEgNEC+SlJSUpKSkpGqf2759e5V1Y8aMsftYI4DmiTc+87m7WTT/nsia75e4J9Jlb4IwE7/bCECjVfnGJ/38RleJNz5zNLb7JWA+4gVAo8YbX9MwvEeIPkuOsS2nTRqofzz1G44fbkqDXDYCgLpoTDcK4uY1lvslYD7OvAAwAm98ACoRLwAAwCjECwAAMArxAgAAjEK8AAAAoxAvAADAKMQLAAAwCvECAACMQrzUUnnFz7+RIzP7nN0yAABoOMRLLWw6mKu4JTtsyxPf/FJD//y5Nh3MdeGsAABonoiX69h0MFfT1uxXflGJ3fq8wsuatmY/AQMAQAMjXmpQXmHVwo8OqboLRJXrFn50iEtIAAA0IOKlBpnZ55RbeNnh81ZJuYWXlZl9ruEmBQBAM0e81ODMBcfhcjPjAABA3REvNbiltU+9jgMAAHVHvNQgKrydQvx9ZHHwvEVSiL+PosLbNeS0AABo1oiXGri7WTT/nkhJqhIwlcvz74mUu5ujvAEAAPWNeLmO4T1CtHJcP93i5223PtjfRyvH9dPwHiEumhkAAM2Th6snYILhPUI0pGuAei74VJKUNmmg7ogI5IwLAAAuwJmXWro6VKLC2xEuAAC4CPECAACMQrwAAACjEC8AAMAoxAsAADAK8QIAAIxCvAAAAKMQLwAAwCjECwAAMArxAgAAjEK8AAAAoxAvAADAKMQLAAAwCvECAACMQrwAAACjEC8AAMAoxAsAADAK8QIAAIxCvAAAAKMQLwAAwCjECwAAMArxAgAAjEK8AAAAoxAvAADAKMQLAAAwCvECAACMQrwAAACjEC8AAMAoxAsAADAK8QIAAIxCvAAAAKMQLwAAwCjECwAAMArxAgAAjEK8AAAAoxAvAADAKMQLAAAwCvECAACMQrwAAACjEC8AAMAoxAsAADCKU+Pl3Llzeuihh+Tn56c2bdpo8uTJunjxYo3bxMbGymKx2D0effRRZ04TAAAYxMOZO3/ooYeUm5urLVu26MqVK5o0aZKmTp2qt956q8btpkyZomeffda27Ovr68xpAgAAgzgtXg4fPqxNmzbpyy+/1IABAyRJr7zyiu6++2699NJL6tChg8NtfX19FRwc7KypAQAAgzntslF6erratGljCxdJiouLk5ubmzIyMmrcdu3atQoICFCPHj00Z84cFRcXOxxbUlKioqIiuwcAAGi6nHbmJS8vT7fccov9F/PwULt27ZSXl+dwuwcffFCdO3dWhw4d9M9//lNPPfWUjh49qvfff7/a8SkpKVq4cGG9zh0AADReNxwvs2fP1p///Ocaxxw+fPimJzR16lTbn3v27KmQkBDdddddOn78uG699dYq4+fMmaPk5GTbclFRkUJDQ2/66wMAgMbthuPl8ccf18SJE2sc06VLFwUHB+vMmTN268vKynTu3Lkbup9l0KBBkqRvvvmm2njx9vaWt7d3rfcHAADMdsPxEhgYqMDAwOuOi46O1vnz57Vv3z71799fkvT555+roqLCFiS1kZWVJUkKCQm50akCAIAmyGk37Hbv3l3Dhw/XlClTlJmZqV27dikpKUkPPPCA7ZNG3333nbp166bMzExJ0vHjx/Xcc89p3759OnHihD788EMlJibq17/+tXr16uWsqQIAAIM49YfUrV27Vt26ddNdd92lu+++W0OHDtWrr75qe/7KlSs6evSo7dNEXl5e+uyzzzRs2DB169ZNjz/+uH7729/qo48+cuY0AQCAQZz6Q+ratWtX4w+kCwsLk9VqtS2HhoZqx44dzpwSAAAwHL/bCAAAGIV4AQAARiFeAACAUYgXAABgFOIFAAAYhXgBAABGIV4AAIBRiBcAAGAU4gUAABiFeAEAAEYhXgAAgFGIFwAAYBTiBQAAGIV4AQAARiFeAACAUYgXAABgFOIFAAAYhXgBAABGIV4AAIBRiBcAAGAU4gUAABiFeAEAAEYhXgAAgFGIFwAAYBTiBQAAGIV4AQAARiFeAACAUYgXAABgFOIFAAAYhXgBAABGIV4AAIBRiBcAAGAU4gUAABiFeAEAAEYhXgAAgFGIFwAAYBTiBQAAGIV4AQAARiFeAACAUYgXAABgFOIFAAAYhXgBAABGIV4AAIBRiBcAAGAU4gUAABiFeAEAAEYhXgAAgFGIFwAAYBTiBQAAGMXD1RMAGoqvl4dOvDDS1dMAANQRZ14AAIBRiBcAAGAU4gUAABiFeAEAAEYhXgAAgFGIFwAAYBTiBQAAGIV4AQAARiFeAACAUYgXAABgFOIFAAAYhXgBAABGIV4AAIBRnBYv//3f/63BgwfL19dXbdq0qdU2VqtV8+bNU0hIiFq0aKG4uDgdO3bMWVMEAAAGclq8lJaWasyYMZo2bVqtt3nxxRf18ssva9WqVcrIyFDLli0VHx+vy5cvO2uaAADAMB7O2vHChQslSWlpabUab7VatWzZMs2dO1cJCQmSpL/97W8KCgrS+vXr9cADDzhrqgAAwCBOi5cblZ2drby8PMXFxdnW+fv7a9CgQUpPT3cYLyUlJSopKbEtFxUVOX2uAAA0R75eHjrxwkhXT6Px3LCbl5cnSQoKCrJbHxQUZHuuOikpKfL397c9QkNDnTpPAADgWjcUL7Nnz5bFYqnxceTIEWfNtVpz5sxRYWGh7XHq1KkG/foAAKBh3dBlo8cff1wTJ06scUyXLl1uaiLBwcGSpPz8fIWEhNjW5+fnq0+fPg638/b2lre39019TQAAYJ4bipfAwEAFBgY6ZSLh4eEKDg7W1q1bbbFSVFSkjIyMG/rEEgCgcWos90vAfE67YTcnJ0fnzp1TTk6OysvLlZWVJUnq2rWrWrVqJUnq1q2bUlJSdN9998lisWjWrFl6/vnnFRERofDwcD3zzDPq0KGDRo0a5axpAjAEb3wAKjktXubNm6fVq1fblvv27StJ2rZtm2JjYyVJR48eVWFhoW3Mk08+qUuXLmnq1Kk6f/68hg4dqk2bNsnHx8dZ0wQAAIaxWK1Wq6snUZ+Kiork7++vwsJC+fn51dt+i0vLFDlvsyTp0LPx8vVqNJ8yBwDAeDfy/t1oPioNAABQG8QLAAAwCvECAACMQrwAAACjEC8AAMAoxAsAADAK8QIAAIxCvAAAAKMQLwAAwCjECwAAMArxAgAAjEK8AAAAoxAvAADAKMQLAAAwCvECAACM4uHqCZjC18tDJ14Y6eppAADQ7HHmBQAAGIV4AQAARiFeAACAUYgXAABgFOIFAAAYhXgBAABGIV4AAIBRiBcAAGAU4gUAABiFeAEAAEYhXgAAgFGIFwAAYBTiBQAAGIV4AQAARiFeAACAUTxcPYH6ZrVaJUlFRUUungkAAKityvftyvfxmjS5eLlw4YIkKTQ01MUzAQAAN+rChQvy9/evcYzFWpvEMUhFRYVOnz6t1q1by2Kx1Ou+i4qKFBoaqlOnTsnPz69e942GwTE0G8fPfBxD8znrGFqtVl24cEEdOnSQm1vNd7U0uTMvbm5u6tSpk1O/hp+fH990huMYmo3jZz6OofmccQyvd8alEjfsAgAAoxAvAADAKMTLDfD29tb8+fPl7e3t6qngJnEMzcbxMx/H0HyN4Rg2uRt2AQBA08aZFwAAYBTiBQAAGIV4AQAARiFeAACAUYgXNHmxsbGaNWuWq6cBNEkTJ07UqFGjXD0NNDNN7ifsAgAaTmpqaq1+kR7MEBsbqz59+mjZsmWunkqNiBcAwE2r7Y9zB+oTl41q6e9//7t69uypFi1aqH379oqLi9OlS5dcPS3UUllZmZKSkuTv76+AgAA988wz/N9iIxMbG6sZM2Zo1qxZatu2rYKCgvTaa6/p0qVLmjRpklq3bq2uXbvqk08+kSSlpaWpTZs2dvtYv359vf9CVvzE0WvgtZeNLly4oIceekgtW7ZUSEiIli5dWuXSbVhYmJ5//nklJiaqVatW6ty5sz788EMVFBQoISFBrVq1Uq9evbR3717bNt9//73Gjh2rjh07ytfXVz179tTbb7/dgP8CTd/EiRO1Y8cOpaamymKxyGKx6Pjx45o8ebLCw8PVokUL3XbbbUpNTbXbrrpL86NGjdLEiROdNlfipRZyc3M1duxYPfzwwzp8+LC2b9+u0aNH8+ZnkNWrV8vDw0OZmZlKTU3VkiVL9Prrr7t6WrjG6tWrFRAQoMzMTM2YMUPTpk3TmDFjNHjwYO3fv1/Dhg3T+PHjVVxc7OqpNis38hqYnJysXbt26cMPP9SWLVu0c+dO7d+/v8q4pUuXasiQITpw4IBGjhyp8ePHKzExUePGjdP+/ft16623KjEx0fY1Ll++rP79+2vjxo06ePCgpk6dqvHjxyszM9Ppf//mIjU1VdHR0ZoyZYpyc3OVm5urTp06qVOnTlq3bp0OHTqkefPm6emnn9Z7773n2slacV379u2zSrKeOHHC1VPBTYiJibF2797dWlFRYVv31FNPWbt37+7CWeFaMTEx1qFDh9qWy8rKrC1btrSOHz/eti43N9cqyZqenm598803rf7+/nb7+OCDD6y8rNW/ml4DJ0yYYE1ISLBarVZrUVGR1dPT07pu3Trb8+fPn7f6+vpaZ86caVvXuXNn67hx42zLlcf1mWeesa1LT0+3SrLm5uY6nNfIkSOtjz/+eB3+ZrhWTEyM3bGqzmOPPWb97W9/W+M2CQkJ1gkTJtT/BP8PZ15qoXfv3rrrrrvUs2dPjRkzRq+99pp++OEHV08LN+BXv/qV3eWE6OhoHTt2TOXl5S6cFa7Vq1cv25/d3d3Vvn179ezZ07YuKChIknTmzJkGn1tzVtvXwG+//VZXrlxRVFSUbZ2/v79uu+22KmOvPtaVx7WmY11eXq7nnntOPXv2VLt27dSqVStt3rxZOTk59fOXhEPLly9X//79FRgYqFatWunVV191+b878VIL7u7u2rJliz755BNFRkbqlVde0W233abs7GxXTw1oUjw9Pe2WLRaL3brKAK2oqJCbm1uVyxZXrlxx/iSbIWe8BlZ3XB0da0latGiRUlNT9dRTT2nbtm3KyspSfHy8SktLb3oOuL533nlHTzzxhCZPnqxPP/1UWVlZmjRpkt2/uyu+F4mXWrJYLBoyZIgWLlyoAwcOyMvLSx988IGrp4VaysjIsFves2ePIiIi5O7u7qIZoa4CAwN14cIFuxvns7KyXDehJq42r4FdunSRp6envvzyS9u6wsJCff3113X++rt27VJCQoLGjRun3r17q0uXLvWyX9jz8vKyOyO9a9cuDR48WNOnT1ffvn3VtWtXHT9+3G6bwMBA5ebm2pbLy8t18OBBp86TeKmFjIwM/elPf9LevXuVk5Oj999/XwUFBerevburp4ZaysnJUXJyso4ePaq3335br7zyimbOnOnqaaEOBg0aJF9fXz399NM6fvy43nrrLaWlpbl6Wk1SbV8DW7durQkTJuiPf/yjtm3bpn/961+aPHmy3Nzc6vwpsIiICG3ZskW7d+/W4cOH9Z//+Z/Kz8+v0z5RVVhYmDIyMnTixAmdPXtWERER2rt3rzZv3qyvv/5azzzzjF2cStJvfvMbbdy4URs3btSRI0c0bdo0nT9/3qnzJF5qwc/PT1988YXuvvtu/fKXv9TcuXO1ePFijRgxwtVTQy0lJibqxx9/VFRUlB577DHNnDlTU6dOdfW0UAft2rXTmjVr9PHHH9s+NrtgwQJXT6tJupHXwCVLlig6Olr/8R//obi4OA0ZMkTdu3eXj49PneYwd+5c9evXT/Hx8YqNjVVwcDA/2dcJnnjiCbm7uysyMlKBgYGKj4/X6NGjdf/992vQoEH6/vvvNX36dLttHn74YU2YMEGJiYmKiYlRly5ddOeddzp1nhbrtReqAACoJ5cuXVLHjh21ePFiTZ482dXTQRPBT9gFANSbAwcO6MiRI4qKilJhYaGeffZZSVJCQoKLZ4amhHgBANSrl156SUePHpWXl5f69++vnTt3KiAgwNXTQhPCZSMAAGAUbtgFAABGIV4AAIBRiBcAAGAU4gUAABiFeAEAAEYhXgAAgFGIFwAAYBTiBQAAGOX/B3io3qW7+kDHAAAAAElFTkSuQmCC", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "ref = None\n", "for bin, values in results.items():\n", " plt.figure()\n", " m = np.mean(values, axis=0) - truth\n", " s = np.std(values, axis=0, ddof=1)\n", " plt.title(f\"{bin=}\")\n", " plt.errorbar(np.arange(len(m)), m / s, 1, fmt=\"o\", label=f\"{bin=}\")\n", " plt.axhline(0, ls=\"--\", color=\"0.5\")\n", " plt.xticks(np.arange(len(m)), [\"s\", \"b\", \"mu\", \"sigma\", \"tau\"]);" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The plots show the bias relative to the standard deviation for each parameter. All results are unbiased, whatever the binning. The bias is not exactly zero, since we used only 100 repetitions, it shrinks further with more. One can observe that the residual bias that is coming from the finite sampling is the same for the unbinned fit and the fits with 100 and 200 bins, which are essentially equivalent." ] } ], "metadata": { "interpreter": { "hash": "bdbf20ff2e92a3ae3002db8b02bd1dd1b287e934c884beb29a73dced9dbd0fa3" }, "kernelspec": { "display_name": "Python 3.8.12 ('venv': venv)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.9.18" }, "orig_nbformat": 4 }, "nbformat": 4, "nbformat_minor": 2 }