{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Hesse and Minos\n",
    "\n",
    "We discuss how uncertainty computation with Hesse and Minos works and how the two approaches differ.\n",
    "\n",
    "iminuit (and C++ MINUIT) offers two major ways of computing parameter uncertainties from the cost function (which must be of least-squares form or negative log-likelihood), HESSE and MINOS. These have different pros and cons, on which we already touched in the basic tutorial. Here we want to go a bit deeper into the pros and cons and also try to reveal a bit of the math behind the two approaches.\n",
    "\n",
    "There are several sources which explain what HESSE and MINOS do, in particular the MINUIT User's Guide (linked from the [iminuit documentation](https://scikit-hep.org/iminuit/about.html)). The mathematical details are covered in F. James, \"Statistical Methods in Experimental Physics\", 2nd edition, World Scientific (2006)."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We need to use a couple of technical terms from mathematical statistics when talk about proven properties. We also use recurring variable names with the same meaning throughout.\n",
    "\n",
    "* $n$ number of [observations](https://en.wikipedia.org/wiki/Sample_%28statistics%29)\n",
    "* $x$ observation from data distribution $f(x; \\theta)$ with parameter $\\theta$ (not an angle!)\n",
    "* $\\hat \\theta$ estimate of $\\theta$ obtained from a sample (the parameter value obtained from the fit)\n",
    "* $\\mathcal L$ likelihood function\n",
    "* $\\mathcal I$ [Fisher information](https://en.wikipedia.org/wiki/Fisher_information)\n",
    "* $E[\\dots]$ expectation value of $\\dots$ over distribution of $x$\n",
    "* $V[\\dots]$ variance of $\\dots$\n",
    "\n",
    "The terms **asymptotic** and **asymptotic limit** refer to inference from an infinite data sample. Mathematical properties of statistical methods are most commonly computed in this limit."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## HESSE\n",
    "\n",
    "HESSE in a nutshell:\n",
    "\n",
    "* Computes approximate covariance matrix $C_{ij}$ for all fitted parameters\n",
    "* Square-root of diagonal elements of $C_{ij}$ corresponds to one standard deviation; to obtain $k$ standard deviations, multiply covariance matrix $C$ by $k^2$\n",
    "* Can be used to form confidence intervals $\\hat \\theta_i \\pm C_{ii}^{1/2}$\n",
    "* Only way to obtain parameter correlations\n",
    "* Comparably fast, requires:\n",
    "    * Numerical computation of all second derivatives of the cost function (Hessian matrix)\n",
    "    * Inverting this matrix\n",
    "* Cannot compute uncertainty for only one parameter if there are several\n",
    "* Approximately computed as by-product of MIGRAD algorithm (usually accurate enough when `strategy` >= 1 is used)\n",
    "\n",
    "The algorithm derives its name from the Hessian matrix of second derivatives which is computed to obtain the uncertainties.\n",
    "\n",
    "### How does it work?\n",
    "\n",
    "The matrix of second derivatives of the cost function is computed for all free parameters and multiplied by 2. The result is inverted and multiplied by `errordef`.\n",
    "\n",
    "MINUIT implements parameter limits by applying a variable transformation and therefore distinguishes internal and external parameter space. The Hessian matrix is computed in the internal parameter space and transformed using the chain rule to external space.\n",
    "\n",
    "### Why does it work?\n",
    "\n",
    "One can prove under general conditions in the asymptotic limit that a parameter estimate $\\hat\\theta$ obtained from the least-squares and the maximum-likelihood methods is normally distributed with minimum variance, given by the [Cramer-Rao lower bound](https://en.wikipedia.org/wiki/Cram%C3%A9r%E2%80%93Rao_bound), which is the minimum for any unbiased estimator (these methods are asymptotically unbiased).\n",
    "\n",
    "$$\n",
    "V(\\hat \\theta) \\underset{n\\rightarrow \\infty}{\\longrightarrow} \\left\\{ n E\\left[\\left( \\frac{\\partial \\ln\\! f(x;\\theta)}{\\partial \\theta} \\right)^2 \\right]_{\\theta = \\hat\\theta} \\right\\}^{-1} = \\left\\{ -n E\\left[\\frac{\\partial^2 \\ln\\! f(x;\\theta)}{\\partial \\theta^2} \\right]_{\\theta = \\hat\\theta} \\right\\}^{-1}\n",
    "$$\n",
    "\n",
    "The expectation here is taken over the data distribution. Since the expectation value is constant, we see that the variance of $\\hat\\theta$ scales goes down with $n^{-1}$ in the asymptotic limit.\n",
    "\n",
    "If the data range is independent of $\\theta$, which we usually assume (but see F. James book for a counter example), we can swap integration over $x$ and differentiation with respect to $\\theta$. Doing this and replacing the expectation with its plug-in estimate, the arithmetic average, we obtain:\n",
    "\n",
    "$$\n",
    "-n E\\left[\\frac{\\partial^2 \\ln\\! f(x;\\theta)}{\\partial \\theta^2} \\right]_{\\theta = \\hat \\theta} = -n \\frac{1}{n} \\sum_i \\frac{\\partial^2 \\ln\\! f(x_i; \\theta)}{\\partial \\theta^2}\\Big\\vert_{\\theta = \\hat \\theta} = \\frac{\\partial^2 \\big(-\\sum_i \\ln\\! f(x_i; \\theta)\\big)}{\\partial \\theta^2}\\Big\\vert_{\\theta = \\hat \\theta}\n",
    "$$\n",
    "\n",
    "We now see that the numerator contains the negative log-likelihood  function that we often plug into iminuit. If there is a vector of parameters $\\hat{\\vec \\theta}$, then this turns into the Hessian matrix of second derivatives.\n",
    "\n",
    "#### A bit of history\n",
    "\n",
    "So what about these factors of $2$ and `errordef` in MINUIT that were mentioned above? These don't appear here. There are historical reasons for those. In the asymptotic limit, the least-squares cost function that corresponds to the log-likelihood is $Q = -2 \\ln\\! \\mathcal{L} - \\text{constants}$. MINUIT was originally developed with least-squares fits in mind, therefore its internal math assumes the $Q$ form. If the second derivatives are computed from $Q$, the constants are removed but the Hessian must be multiplied by a factor of two to get the right variance. Correspondingly, if the user puts in a negative log-likelihood function, the same procedure now introduces an extra factor of two $2$, which must be compensated by the `errordef` value of 0.5 for the negative log-likelihood.\n",
    "\n",
    "### Why is HESSE approximate\n",
    "\n",
    "We started out from Cramer-Rao bound, the asymptotic limit for the parameter variance. How fast the finite sample approaches the limit depends on the problem at hand. For normal distributed data, the bound is exact.\n",
    "\n",
    "We further approximated the computation of the bound by replacing the expectation over the likelihood with the sample mean of the likelihood. \n",
    "\n",
    "## MINOS\n",
    "\n",
    "MINOS in a nutshell:\n",
    "\n",
    "* Approximate confidence intervals (intervals are wrongly claimed to be \"exact\" in some sources, including the MINUIT paper from 1975)\n",
    "* Cannot compute parameter correlations\n",
    "* Some (unverified) sources claim better coverage probability than intervals based on HESSE\n",
    "* In general slower than HESSE (requiring more function evaluations):\n",
    "    * Iteratively computes the value pair $\\theta^d, \\theta^u$ which increases the cost function by `errordef` over the minimum\n",
    "    * If the cost function has several parameters, it is minimised with respect to all other parameters during this scan\n",
    "* Can be used to compute uncertainty for only one parameter - but this is not more efficient than HESSE, since the computation requires at least one evaluation of HESSE\n",
    "\n",
    "The MINOS algorithm computes [profile-likelihood based](https://en.wikipedia.org/wiki/Likelihood_function#Profile_likelihood) confidence intervals.\n",
    "\n",
    "### How does it work?\n",
    "\n",
    "MINOS scans the likelihood along one parameter $\\theta_i$,  while minimizing the likelihood with respect to all other parameters $\\theta_k$ with $k \\ne i$. This is effectively the same as expressing the other parameter estimates as a function of $\\theta_i$, $\\hat \\theta_k(\\theta_i)$, and scanning the now one-dimensional negative log-likelihood $-\\ln \\mathcal{L}(\\theta_i; \\theta_k = \\hat \\theta_k(\\theta_i) , k\\ne i)$. This is called the [profile likelihood](https://en.wikipedia.org/wiki/Likelihood_function#Profile_likelihood) for parameter $\\theta_i$.\n",
    "\n",
    "One follows this curve until $-\\ln \\mathcal{L}$ increases by `errordef` with respect to its minimum and stores the two corresponding values $\\theta^d_i$ and $\\theta^u_i$. The interval $(\\theta^d_i, \\theta^u_i)$ has $68\\,\\%$ coverage probability in the asymptotic limit. In multi-dimensional parameter space, the computation is comparably expensive due to the iterative steps of scanning and minimization.\n",
    "\n",
    "An efficient algorithm to compute the interval is described by Venzon and Moolgavkar in *A Method for Computing Profile-Likelihood-Based Confidence Intervals*, Journal of the Royal Statistical Society C37 (1988) 87–94 [DOI](https://doi.org/10.2307%2F2347496).\n",
    "\n",
    "### Why does it work?\n",
    "\n",
    "We define the likelihood ratio $\\ell(\\vec\\theta) = \\mathcal{L}(\\vec\\theta) / \\mathcal{L}(\\hat{\\vec\\theta})$. In the asymptotic limit, $-2 \\ln \\ell(\\vec\\theta)$ is $\\chi^2(k)$ distributed, where $k$ is the number of fitted parameters. For $k = 1$, the $\\chi^2$-interval $[0, 1)$ contains $68\\,\\%$ probability. For a single parameter fit therefore a corresponding interval is found by the two values $\\theta^d$ and $\\theta^u$ which solve\n",
    "\n",
    "$$\n",
    "-2\\ln\\ell(\\theta) = 1 \\Leftrightarrow -\\ln\\ell(\\theta) = 1/2\n",
    "$$\n",
    "\n",
    "We recognize the difference in the negative log-likelihood on the left-hand side and our `errordef = 0.5` on the right-hand side. Confidence intervals with other coverage probability can be constructed by finding the corresponding upper value of the $\\chi^2$-interval with that integrated probability. In a least-squares fit, we have\n",
    "\n",
    "$$\n",
    "\\Delta Q(\\theta) = -2 \\ln \\ell(\\theta) = 1\n",
    "$$\n",
    "\n",
    "therefore `errordef = 1` is the crossing value for a least-squares cost function.\n",
    "\n",
    "In the multi-parameter case, when we search for the interval for $\\theta_i$, the other parameters are effectively fixed to their best-fit values for the current value $\\theta_i$ in the scan, $\\theta_k = \\hat\\theta_k(\\theta_i)$. Therefore, during the scan they are not free. The profile likelihood is effectively a likelihood for a single parameter. Thus, $68\\,\\%$-intervals are also here obtained when the negative log-likelihood crosses the value $1/2$ above the minimum. \n",
    "\n",
    "### Why is MINOS approximate\n",
    "\n",
    "In some sources, the MINOS intervals are wrongly described as *exact*. They are not exact, because for finite samples the intervals do not necessarily have $68\\,\\%$ coverage probability, only in the asymptotic limit.\n",
    "\n",
    "Some sources claim that two approximations are involved in the HESSE calculation of an interval, but only one in the MINOS calculation and conclude that MINOS intervals therefore approach the asymptotic limit more rapidly. This claim is disputed by others."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Coverage probability\n",
    "\n",
    "We compute the coverage probability of HESSE and MINOS intervals in toy experiments.\n",
    "\n",
    "### Poisson distributed data\n",
    "\n",
    "We construct HESSE and MINOS intervals for a counting experiment. We consider the extreme case of a single observation $k$ drawn from a Poisson distribution $P(k;\\lambda)$. We use the maximum-likelihood method to find the best estimate for $\\lambda$ for each $k$ under the constraint $\\lambda > 0$, which is trivially just $\\hat \\lambda = k$, and construct intervals with the HESSE and MINOS algorithms to check their coverage.\n",
    "\n",
    "This case can be fully handled analytically, but here we use iminuit's HESSE and MINOS algorithms to compute the intervals."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from matplotlib import pyplot as plt\n",
    "from scipy.stats import multivariate_normal, poisson\n",
    "from argparse import Namespace\n",
    "from iminuit import Minuit\n",
    "from functools import lru_cache\n",
    "import joblib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "@lru_cache(10000)\n",
    "def run(k):\n",
    "    m = Minuit(lambda lambd: -poisson.logpmf(k, lambd), lambd=k + 1)\n",
    "    m.limits[\"lambd\"] = (0, None)\n",
    "    m.errordef = Minuit.LIKELIHOOD\n",
    "    m.migrad()\n",
    "    m.hesse()\n",
    "    m.minos()\n",
    "    assert m.valid\n",
    "\n",
    "    p = m.values[\"lambd\"]\n",
    "    dp = m.errors[\"lambd\"]\n",
    "    pm = max(p + m.merrors[\"lambd\"].lower, 0.0), p + m.merrors[\"lambd\"].upper\n",
    "\n",
    "    r = p, dp, *pm\n",
    "    return r\n",
    "\n",
    "\n",
    "rng = np.random.default_rng(seed=1)\n",
    "nmc = 5000\n",
    "mu = 10 ** np.linspace(-1, 2, 100)\n",
    "\n",
    "pcov = {\n",
    "    \"HESSE\": np.empty_like(mu),\n",
    "    \"MINOS\": np.empty_like(mu),\n",
    "}\n",
    "\n",
    "for i, mui in enumerate(mu):\n",
    "\n",
    "    nh = 0\n",
    "    nm = 0\n",
    "    for imc in range(nmc):\n",
    "        k = rng.poisson(mui)\n",
    "\n",
    "        p, dp, pd, pu = run(k)\n",
    "\n",
    "        if p - dp < mui < p + dp:\n",
    "            nh += 1\n",
    "        if pd < mui < pu:\n",
    "            nm += 1\n",
    "\n",
    "    pcov[\"HESSE\"][i] = nh / nmc\n",
    "    pcov[\"MINOS\"][i] = nm / nmc\n",
    "\n",
    "fig, ax = plt.subplots(1, 2, figsize=(10, 4))\n",
    "\n",
    "plt.sca(ax[0])\n",
    "n = np.arange(101)\n",
    "interval = {\n",
    "    \"HESSE\": np.empty((len(n), 2)),\n",
    "    \"MINOS\": np.empty((len(n), 2)),\n",
    "}\n",
    "for i, k in enumerate(n):\n",
    "    p, dp, pd, pu = run(k)\n",
    "    interval[\"HESSE\"][i] = (p - dp, p + dp)\n",
    "    interval[\"MINOS\"][i] = (pd, pu)\n",
    "\n",
    "for algo, vals in interval.items():\n",
    "    plt.plot(n, vals[:, 0] - n, color=\"C0\" if algo == \"HESSE\" else \"C1\", label=algo)\n",
    "    plt.plot(n, vals[:, 1] - n, color=\"C0\" if algo == \"HESSE\" else \"C1\", label=algo)\n",
    "plt.xlabel(\"$k$\")\n",
    "plt.ylabel(r\"$\\lambda^d - \\hat\\lambda$, $\\lambda^u - \\hat\\lambda$\")\n",
    "\n",
    "plt.sca(ax[1])\n",
    "for algo, vals in pcov.items():\n",
    "    plt.plot(mu, vals, label=algo)\n",
    "\n",
    "plt.axhline(0.68, ls=\":\", color=\"0.5\", zorder=0)\n",
    "plt.xlabel(r\"$\\lambda$\")\n",
    "plt.ylabel(\"coverage probability\")\n",
    "plt.legend()\n",
    "plt.semilogx();"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this special case, the intervals found by both methods are very close. The MINOS interval is identical to the HESSE interval up to a small almost constant shift. Visually, the rate of converge to the asymptotic coverage probability of 68% seems to be equal for both methods."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can speak about the rate of convergence although we have drawn only a single observation from the Poisson pdf. The log-likelihood for $n$ observations with the same expectation is identical to the log-likelihood for one observation with an $n$-times higher expectation up to additive constants.\n",
    "\n",
    "$$\n",
    "\\ln L = \\sum_i^n (\\lambda_i - k_i \\ln \\lambda_i)\n",
    "= \\lambda - \\sum_i k_i (\\ln \\lambda - \\ln n)\n",
    "= \\lambda - k \\ln \\lambda + k \\ln n\n",
    "$$\n",
    "\n",
    "with $\\sum_i^n \\lambda_i = \\lambda$, $\\sum_i^n k_i = k$, and $\\lambda_i = \\lambda / n$. Therefore, the test cases with large values of $\\lambda$ correspond to large observation samples with a small constant $\\lambda$."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Fit of transformed normally distributed data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "    <tr>\n",
       "        <th colspan=\"5\" style=\"text-align:center\" title=\"Minimizer\"> Migrad </th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td colspan=\"2\" style=\"text-align:left\" title=\"Minimum value of function\"> FCN = 3.122 </td>\n",
       "        <td colspan=\"3\" style=\"text-align:center\" title=\"Total number of function and (optional) gradient evaluations\"> Nfcn = 168 </td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td colspan=\"2\" style=\"text-align:left\" title=\"Estimated distance to minimum and goal\"> EDM = 2.02e-05 (Goal: 0.0001) </td>\n",
       "        <td colspan=\"3\" style=\"text-align:center\" title=\"Total run time of algorithms\">  </td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td colspan=\"2\" style=\"text-align:center;background-color:#92CCA6;color:black\"> Valid Minimum </td>\n",
       "        <td colspan=\"3\" style=\"text-align:center;background-color:#92CCA6;color:black\"> No Parameters at limit </td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td colspan=\"2\" style=\"text-align:center;background-color:#92CCA6;color:black\"> Below EDM threshold (goal x 10) </td>\n",
       "        <td colspan=\"3\" style=\"text-align:center;background-color:#92CCA6;color:black\"> Below call limit </td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <td style=\"text-align:center;background-color:#92CCA6;color:black\"> Covariance </td>\n",
       "        <td style=\"text-align:center;background-color:#92CCA6;color:black\"> Hesse ok </td>\n",
       "        <td style=\"text-align:center;background-color:#92CCA6;color:black\" title=\"Is covariance matrix accurate?\"> Accurate </td>\n",
       "        <td style=\"text-align:center;background-color:#92CCA6;color:black\" title=\"Is covariance matrix positive definite?\"> Pos. def. </td>\n",
       "        <td style=\"text-align:center;background-color:#92CCA6;color:black\" title=\"Was positive definiteness enforced by Minuit?\"> Not forced </td>\n",
       "    </tr>\n",
       "</table>"
      ],
      "text/plain": [
       "┌─────────────────────────────────────────────────────────────────────────┐\n",
       "│                                Migrad                                   │\n",
       "├──────────────────────────────────┬──────────────────────────────────────┤\n",
       "│ FCN = 3.122                      │              Nfcn = 168              │\n",
       "│ EDM = 2.02e-05 (Goal: 0.0001)    │                                      │\n",
       "├──────────────────────────────────┼──────────────────────────────────────┤\n",
       "│          Valid Minimum           │        No Parameters at limit        │\n",
       "├──────────────────────────────────┼──────────────────────────────────────┤\n",
       "│ Below EDM threshold (goal x 10)  │           Below call limit           │\n",
       "├───────────────┬──────────────────┼───────────┬─────────────┬────────────┤\n",
       "│  Covariance   │     Hesse ok     │ Accurate  │  Pos. def.  │ Not forced │\n",
       "└───────────────┴──────────────────┴───────────┴─────────────┴────────────┘"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table>\n",
       "    <tr>\n",
       "        <td></td>\n",
       "        <th title=\"Variable name\"> Name </th>\n",
       "        <th title=\"Value of parameter\"> Value </th>\n",
       "        <th title=\"Hesse error\"> Hesse Error </th>\n",
       "        <th title=\"Minos lower error\"> Minos Error- </th>\n",
       "        <th title=\"Minos upper error\"> Minos Error+ </th>\n",
       "        <th title=\"Lower limit of the parameter\"> Limit- </th>\n",
       "        <th title=\"Upper limit of the parameter\"> Limit+ </th>\n",
       "        <th title=\"Is the parameter fixed in the fit\"> Fixed </th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <th> 0 </th>\n",
       "        <td> cx </td>\n",
       "        <td> 0.11 </td>\n",
       "        <td> 0.06 </td>\n",
       "        <td> -0.08 </td>\n",
       "        <td> 0.05 </td>\n",
       "        <td>  </td>\n",
       "        <td>  </td>\n",
       "        <td>  </td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <th> 1 </th>\n",
       "        <td> cy </td>\n",
       "        <td> 0.05 </td>\n",
       "        <td> 0.10 </td>\n",
       "        <td> -0.11 </td>\n",
       "        <td> 0.08 </td>\n",
       "        <td>  </td>\n",
       "        <td>  </td>\n",
       "        <td>  </td>\n",
       "    </tr>\n",
       "</table>"
      ],
      "text/plain": [
       "┌───┬──────┬───────────┬───────────┬────────────┬────────────┬─────────┬─────────┬───────┐\n",
       "│   │ Name │   Value   │ Hesse Err │ Minos Err- │ Minos Err+ │ Limit-  │ Limit+  │ Fixed │\n",
       "├───┼──────┼───────────┼───────────┼────────────┼────────────┼─────────┼─────────┼───────┤\n",
       "│ 0 │ cx   │   0.11    │   0.06    │   -0.08    │    0.05    │         │         │       │\n",
       "│ 1 │ cy   │   0.05    │   0.10    │   -0.11    │    0.08    │         │         │       │\n",
       "└───┴──────┴───────────┴───────────┴────────────┴────────────┴─────────┴─────────┴───────┘"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table>\n",
       "    <tr>\n",
       "        <td></td>\n",
       "        <th colspan=\"2\" style=\"text-align:center\" title=\"Parameter name\"> cx </th>\n",
       "        <th colspan=\"2\" style=\"text-align:center\" title=\"Parameter name\"> cy </th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <th title=\"Lower and upper minos error of the parameter\"> Error </th>\n",
       "        <td> -0.08 </td>\n",
       "        <td> 0.05 </td>\n",
       "        <td> -0.11 </td>\n",
       "        <td> 0.08 </td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <th title=\"Validity of lower/upper minos error\"> Valid </th>\n",
       "        <td style=\"background-color:#92CCA6;color:black\"> True </td>\n",
       "        <td style=\"background-color:#92CCA6;color:black\"> True </td>\n",
       "        <td style=\"background-color:#92CCA6;color:black\"> True </td>\n",
       "        <td style=\"background-color:#92CCA6;color:black\"> True </td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <th title=\"Did scan hit limit of any parameter?\"> At Limit </th>\n",
       "        <td style=\"background-color:#92CCA6;color:black\"> False </td>\n",
       "        <td style=\"background-color:#92CCA6;color:black\"> False </td>\n",
       "        <td style=\"background-color:#92CCA6;color:black\"> False </td>\n",
       "        <td style=\"background-color:#92CCA6;color:black\"> False </td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <th title=\"Did scan hit function call limit?\"> Max FCN </th>\n",
       "        <td style=\"background-color:#92CCA6;color:black\"> False </td>\n",
       "        <td style=\"background-color:#92CCA6;color:black\"> False </td>\n",
       "        <td style=\"background-color:#92CCA6;color:black\"> False </td>\n",
       "        <td style=\"background-color:#92CCA6;color:black\"> False </td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "        <th title=\"New minimum found when doing scan?\"> New Min </th>\n",
       "        <td style=\"background-color:#92CCA6;color:black\"> False </td>\n",
       "        <td style=\"background-color:#92CCA6;color:black\"> False </td>\n",
       "        <td style=\"background-color:#92CCA6;color:black\"> False </td>\n",
       "        <td style=\"background-color:#92CCA6;color:black\"> False </td>\n",
       "    </tr>\n",
       "</table>"
      ],
      "text/plain": [
       "┌──────────┬───────────────────────┬───────────────────────┐\n",
       "│          │          cx           │          cy           │\n",
       "├──────────┼───────────┬───────────┼───────────┬───────────┤\n",
       "│  Error   │   -0.08   │   0.05    │   -0.11   │   0.08    │\n",
       "│  Valid   │   True    │   True    │   True    │   True    │\n",
       "│ At Limit │   False   │   False   │   False   │   False   │\n",
       "│ Max FCN  │   False   │   False   │   False   │   False   │\n",
       "│ New Min  │   False   │   False   │   False   │   False   │\n",
       "└──────────┴───────────┴───────────┴───────────┴───────────┘"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "(-0.1, 0.2)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1400x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "rng = np.random.default_rng(1)\n",
    "\n",
    "truth = Namespace(cr=0.1, cphi=0, sr=0.1, sphi=2)\n",
    "truth.cx = truth.cr * np.cos(truth.cphi)\n",
    "truth.cy = truth.cr * np.sin(truth.cphi)\n",
    "\n",
    "d_r = rng.normal(truth.cr, truth.sr, size=5)\n",
    "d_phi = rng.normal(truth.cphi, truth.sphi, size=d_r.shape)\n",
    "\n",
    "cov = np.eye(2)\n",
    "cov[0, 0] = truth.sr ** 2\n",
    "cov[1, 1] = truth.sphi ** 2\n",
    "\n",
    "\n",
    "def nll(cx, cy):\n",
    "    cr = np.linalg.norm((cx, cy))\n",
    "    cphi = np.arctan2(cy, cx)\n",
    "    return -np.sum(\n",
    "        multivariate_normal((cr, cphi), cov).logpdf(np.transpose((d_r, d_phi)))\n",
    "    )\n",
    "\n",
    "\n",
    "m = Minuit(nll, cx=0.1, cy=0)\n",
    "m.errordef = Minuit.LIKELIHOOD\n",
    "m.migrad()\n",
    "m.hesse()\n",
    "m.minos()\n",
    "display(m.fmin, m.params)\n",
    "display(m.merrors)\n",
    "\n",
    "plt.figure(figsize=(14, 4))\n",
    "\n",
    "plt.subplot(121, polar=True)\n",
    "plt.plot(d_phi, d_r, \".\")\n",
    "\n",
    "plt.subplot(122, aspect=\"equal\")\n",
    "m.draw_mncontour(\"cx\", \"cy\", size=100)\n",
    "plt.plot(m.values[\"cx\"], m.values[\"cy\"], \"+k\")\n",
    "plt.xlim(-0.1, 0.2)\n",
    "plt.ylim(-0.1, 0.2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "truth = Namespace(cr=0.1, cphi=0, sr=0.1, sphi=2)\n",
    "truth.cx = truth.cr * np.cos(truth.cphi)\n",
    "truth.cy = truth.cr * np.sin(truth.cphi)\n",
    "truth.cov = np.eye(2)\n",
    "truth.cov[0, 0] = truth.sr ** 2\n",
    "truth.cov[1, 1] = truth.sphi ** 2\n",
    "\n",
    "n_tries = 500  # increase this to 500 get smoother curves (running will take a while)\n",
    "n_data = np.unique(np.geomspace(5, 100, 20, dtype=int))\n",
    "\n",
    "\n",
    "@joblib.delayed\n",
    "def run(n):\n",
    "    rng = np.random.default_rng(seed=n)\n",
    "\n",
    "    n_h = 0\n",
    "    n_m = 0\n",
    "    h_lus = []\n",
    "    m_lus = []\n",
    "    xs = []\n",
    "    for i_try in range(n_tries):\n",
    "        while True:\n",
    "            d_r = rng.normal(truth.cr, truth.sr, size=n)\n",
    "            d_phi = rng.normal(truth.cphi, truth.sphi, size=n)\n",
    "\n",
    "            def nll(cx, cy):\n",
    "                cr = np.linalg.norm((cx, cy))\n",
    "                cphi = np.arctan2(cy, cx)\n",
    "                return -np.sum(\n",
    "                    multivariate_normal((cr, cphi), truth.cov).logpdf(\n",
    "                        np.transpose((d_r, d_phi))\n",
    "                    )\n",
    "                )\n",
    "\n",
    "            m = Minuit(nll, cx=0.1, cy=0)\n",
    "            m.errordef = Minuit.LIKELIHOOD\n",
    "            try:\n",
    "                m.migrad()\n",
    "                if not m.valid:\n",
    "                    continue\n",
    "\n",
    "                m.hesse()\n",
    "\n",
    "                if not m.accurate:\n",
    "                    continue\n",
    "\n",
    "                m.minos(\"cx\")\n",
    "                if m.merrors[\"cx\"].is_valid:\n",
    "                    break\n",
    "\n",
    "            except Exception as e:\n",
    "                print(f\"exception in n={n} i_try={i_try}\")\n",
    "                print(e)\n",
    "\n",
    "        x = m.values[\"cx\"]\n",
    "        dx = m.errors[\"cx\"]\n",
    "        me = m.merrors[\"cx\"]\n",
    "        h_lu = x - dx, x + dx\n",
    "        m_lu = x + me.lower, x + me.upper\n",
    "        if h_lu[0] < truth.cx < h_lu[1]:\n",
    "            n_h += 1\n",
    "        if m_lu[0] < truth.cx < m_lu[1]:\n",
    "            n_m += 1\n",
    "        xs.append(x)\n",
    "        h_lus.append(h_lu)\n",
    "        m_lus.append(m_lu)\n",
    "\n",
    "    x = np.mean(xs)\n",
    "    h_l, h_u = np.mean(h_lus, axis=0)\n",
    "    m_l, m_u = np.mean(m_lus, axis=0)\n",
    "    return n_h, n_m, x, h_l, h_u, m_l, m_u\n",
    "\n",
    "\n",
    "n_h, n_m, x, hl, hu, ml, mu = np.transpose(joblib.Parallel(-1)(run(x) for x in n_data))\n",
    "\n",
    "h_pcov = n_h.astype(float) / n_tries\n",
    "m_pcov = n_m.astype(float) / n_tries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(1, 2, figsize=(10, 4))\n",
    "plt.sca(ax[0])\n",
    "plt.fill_between(n_data, hl, hu, alpha=0.5, label=\"HESSE\")\n",
    "plt.fill_between(n_data, ml, mu, alpha=0.5, label=\"MINOS\")\n",
    "plt.plot(n_data, x, \"-k\")\n",
    "plt.legend()\n",
    "plt.xlabel(\"n\")\n",
    "plt.ylabel(\"cx\")\n",
    "plt.semilogx()\n",
    "plt.sca(ax[1])\n",
    "plt.plot(n_data, h_pcov, label=\"HESSE\")\n",
    "plt.plot(n_data, m_pcov, label=\"MINOS\")\n",
    "plt.axhline(0.68, ls=\":\", color=\"0.5\", zorder=0)\n",
    "plt.xlabel(r\"cx\")\n",
    "plt.ylabel(\"coverage probability\")\n",
    "plt.legend()\n",
    "plt.ylim(0, 1)\n",
    "plt.semilogx();"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.8.14 ('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": "3.10.8 (main, Oct 13 2022, 09:48:40) [Clang 14.0.0 (clang-1400.0.29.102)]"
  },
  "vscode": {
   "interpreter": {
    "hash": "bdbf20ff2e92a3ae3002db8b02bd1dd1b287e934c884beb29a73dced9dbd0fa3"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}