{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "7f310032-695b-4be0-8dce-6da3822fea9a",
   "metadata": {},
   "outputs": [],
   "source": [
    "\"\"\"\n",
    "Confidence intervals for a mean from N measurements when sigma is unknown.\n",
    "\n",
    "Recipe A (wrong for small N):  xbar +/- 1.96      * s/sqrt(N)\n",
    "Recipe B (correct):            xbar +/- t_{N-1}   * s/sqrt(N)\n",
    "\n",
    "\"Coverage\" = fraction of repeated experiments in which the interval\n",
    "contains the TRUE mean mu. It is a property of the recipe, not of one interval.\n",
    "\"\"\"\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy import stats\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "ac76f27a-dc49-4c68-b0b7-8b688d6d07bb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "xbar = 10.04, s = 0.59 (scatter of single points), s/sqrt(N) = 0.26\n",
      "Gaussian interval: 10.04 +/- 0.51\n",
      "t interval:        10.04 +/- 0.73   (t_4 = 2.78)\n",
      "Note: the interval is for the MEAN; single points scatter by s, so several\n",
      "of the 5 points lying outside it is expected.\n",
      "\n"
     ]
    }
   ],
   "source": [
    "\n",
    "rng = np.random.default_rng(1)\n",
    "mu, sigma = 10.0, 0.6          # true flux and true scatter (unknown to the observer)\n",
    "\n",
    "# ---------- 1) Worked example: one data set of 5 measurements ----------\n",
    "x = np.array([10.2, 9.6, 10.9, 10.1, 9.4])\n",
    "N = len(x)\n",
    "xbar, s = x.mean(), x.std(ddof=1)\n",
    "se = s / np.sqrt(N)                      # standard error of the MEAN\n",
    "z95, t95 = 1.96, stats.t.ppf(0.975, N - 1)\n",
    "print(f\"xbar = {xbar:.2f}, s = {s:.2f} (scatter of single points), s/sqrt(N) = {se:.2f}\")\n",
    "print(f\"Gaussian interval: {xbar:.2f} +/- {z95*se:.2f}\")\n",
    "print(f\"t interval:        {xbar:.2f} +/- {t95*se:.2f}   (t_4 = {t95:.2f})\")\n",
    "print(\"Note: the interval is for the MEAN; single points scatter by s, so several\")\n",
    "print(\"of the 5 points lying outside it is expected.\\n\")\n",
    "\n",
    "# ---------- 2) Forty repeated experiments, N = 5 each ----------\n",
    "# Expected misses: 12% of 40 = 4.9 for 1.96, 5% of 40 = 2 for t.\n",
    "n_exp = 40\n",
    "data = np.random.default_rng(3).normal(mu, sigma, (n_exp, N)) # 3 is a random seed\n",
    "m = data.mean(axis=1)\n",
    "se_k = data.std(axis=1, ddof=1) / np.sqrt(N)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "d328e156-c862-454f-93d4-cd8f39c00b94",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1100x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(1, 2, figsize=(11, 6), sharey=True, sharex=True)\n",
    "for ax, k, title in zip(axes, [z95, t95], [\"1.96 · s/√N\", f\"t₄ = {t95:.2f} · s/√N\"]):\n",
    "    lo, hi = m - k * se_k, m + k * se_k\n",
    "    miss = (lo > mu) | (hi < mu)\n",
    "    for i in range(n_exp):\n",
    "        ax.plot([lo[i], hi[i]], [i, i], color=\"tab:red\" if miss[i] else \"0.3\", lw=2.5)\n",
    "        ax.plot(m[i], i, \"o\", color=\"tab:red\" if miss[i] else \"0.3\", ms=4)\n",
    "    ax.axvline(mu, color=\"tab:orange\", lw=1.5, label=\"true mean μ\")\n",
    "    ax.set_title(f\"{title}: {miss.sum()} of {n_exp} miss μ\")\n",
    "    ax.set_xlabel(\"flux\")\n",
    "axes[0].set_ylabel(\"experiment number\")\n",
    "axes[0].legend(frameon=False, loc=\"lower left\")\n",
    "fig.tight_layout()\n",
    "fig.savefig(\"t_intervals.png\", dpi=150)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "3f979270-5f6b-4fcd-b815-49512b179618",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "N =  2: coverage 1.96 -> 0.701,  t -> 0.950\n",
      "N =  5: coverage 1.96 -> 0.879,  t -> 0.951\n",
      "N = 10: coverage 1.96 -> 0.919,  t -> 0.950\n",
      "N = 30: coverage 1.96 -> 0.940,  t -> 0.950\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 650x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ---------- 3) Coverage vs N, from many repeated experiments ----------\n",
    "n_rep = 100_000\n",
    "Ns = np.arange(2, 31)\n",
    "cov_z, cov_t = [], []\n",
    "for n in Ns:\n",
    "    d = rng.normal(mu, sigma, (n_rep, n))\n",
    "    mm = d.mean(axis=1)\n",
    "    ee = d.std(axis=1, ddof=1) / np.sqrt(n)\n",
    "    cov_z.append(np.mean(np.abs(mm - mu) < 1.96 * ee))\n",
    "    cov_t.append(np.mean(np.abs(mm - mu) < stats.t.ppf(0.975, n - 1) * ee))\n",
    "    if n in (2, 5, 10, 30):\n",
    "        print(f\"N = {n:2d}: coverage 1.96 -> {cov_z[-1]:.3f},  t -> {cov_t[-1]:.3f}\")\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(6.5, 4))\n",
    "ax.plot(Ns, 100 * np.array(cov_t), \"o-\", label=\"t interval\")\n",
    "ax.plot(Ns, 100 * np.array(cov_z), \"o-\", label=\"1.96 interval\")\n",
    "ax.axhline(95, ls=\":\", color=\"gray\")\n",
    "ax.set_xlabel(\"number of measurements N\")\n",
    "ax.set_ylabel(\"experiments whose interval contains μ (%)\")\n",
    "ax.legend(frameon=False)\n",
    "fig.tight_layout()\n",
    "fig.savefig(\"t_coverage.png\", dpi=150)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "342ee461-c6f8-4d31-84e3-16469be7be73",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}
