Added use of the scipy describe function

Showed how to use describe() get the summary statistics for
a sampled distribution.

03-Gaussians.ipynb

@@ -463,9 +463,7 @@
   {
    "cell_type": "code",
    "execution_count": 4,
-   "metadata": {
-    "collapsed": true
-   },
+   "metadata": {},
    "outputs": [],
    "source": [
     "X = [1.8, 2.0, 1.7, 1.9, 1.6]\n",
@@ -540,7 +538,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "0.02 meters squared\n"
+      "0.01999999999999999 meters squared\n"
      ]
     }
    ],
@@ -609,9 +607,9 @@
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAwUAAAEvCAYAAAAUzW0AAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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 truncated
+      "image/png": "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 truncated
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x26fb1488128>"
+       "<matplotlib.figure.Figure at 0x2c2130495c0>"
       ]
      },
      "metadata": {},
@@ -642,9 +640,9 @@
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAwgAAAEvCAYAAADhM+awAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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 truncated
+      "image/png": "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 truncated
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x26fb34ede48>"
+       "<matplotlib.figure.Figure at 0x2c214170550>"
       ]
      },
      "metadata": {},
@@ -654,8 +652,8 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "mean = 1.795\n",
-      "std  = 0.145\n"
+      "mean = 1.776\n",
+      "std  = 0.147\n"
      ]
     }
    ],
@@ -766,9 +764,9 @@
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAukAAAEHCAYAAAATYthkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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 truncated
+      "image/png": "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\n",
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x26fb3762550>"
+       "<matplotlib.figure.Figure at 0x2c21442bc50>"
       ]
      },
      "metadata": {},
@@ -845,9 +843,9 @@
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAv8AAAEvCAYAAADSLa6QAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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 truncated
+      "image/png": "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 truncated
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x26fb34e6438>"
+       "<matplotlib.figure.Figure at 0x2c214427a58>"
       ]
      },
      "metadata": {},
@@ -884,9 +882,9 @@
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAusAAAEgCAYAAAANPj3XAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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 truncated
+      "image/png": "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 truncated
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x26fb47f2ba8>"
+       "<matplotlib.figure.Figure at 0x2c213f2e6d8>"
       ]
      },
      "metadata": {},
@@ -922,9 +920,9 @@
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAwAAAAEvCAYAAADy5KZEAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAAPYQAAD2EBqD+naQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzs3Xlc1NX+x/HXDMMiiwsgIu4rGiK4\nIK6ldS1tMy1TywzUFiur203zV/dm3Tar2715TSvNlDK1cskWl9LMzF1TFNxRcBcFlU2WWX5/kFxJ\nVEbRLzDv5+Phg0dfvmfmPXQYzme+53uOyeFwOBAREREREZdgNjqAiIiIiIhcPyoARERERERciAoA\nEREREREXogJARERERMSFqAAQEREREXEhKgBERERERFyIxegA15vdbsdutxc7ZjKZMJlMBiUSERER\nEblyDoeDP6/sbzabMZtL/qzfJQuA7Oxso2OIiIiIiFwzPj4+Fy0ANAVIRERERMSFqAAQEREREXEh\nKgBERERERFyIy90DUNLNvpeaIyWXl5iYiNVqxWKxEBYWZnQcMZj6g/yZ+oScT/1B/kx94uqVdI/r\npRa4UQHApe+Slsuz2+3YbDb9HAVQf5ALqU/I+dQf5M/UJ66NSxUA+imLiIiIiLgQFQAiIiIiIi5E\nBYCIiIiIiAtRASAiIiIi4kJUAIiIiIiIuBCXWwVIRETKnsPh4HROASez8jiRmUdGbgF5Vjt5BXb2\npWSBw0YVdwuHzUfx8bRQ09eTQD8PAnw8cTNffKUKEREpeyoARESk1BwOB4dOnWXLwdPsOJrBvhPZ\n7DuZRXJaDvlW++UfYO2pYv9pNkHdGt40rulD40BfQoN9iahXnWZBfioMRESuERUAIiJyUQ6Hg6QT\nWfy6+ySrk06y+cBp0rLzL3p+tSru1PTzpHoVdzzdzXha3MjJysBqs2N1mPDw8iEjt/BKQVp2PnYH\nHEjP4UB6Dr/sOlH0ON4eboTXqUbnJoF0bRZIRN1qWNw0a1VEpCyoABARkWKsNjvr9qfzw7aj/Lwj\nlWMZucW+7+5m4obaVQmrU40mNX1pXNOHJoG+1KrmiafF7YLHi4+Pp6CgAHd3dyIiIoo9z8msfPaf\nLLyKsO9ENtuPZLD10Gmy822s25/Ouv3p/Gfpbvy8LNzYrCa9w4O5uUUQ3h768yUicqX0DioiIjgc\nDrYcPM3Xmw6xOOEY6ed9yu9hMRPdyJ+uTQPp0MiflrWr4uV+4UDfWRY3M8HVvAiu5kWnJgFFx232\nwqsOG5NPsWrvSX7be5IzZwv4YdtRfth2FC93Mz1Cg7i3bV26h9bUlQERESepABARcWFncgr4Zsth\nZq0/wM5jmUXHa3i7c1tYML1aBdOxcUCZDPhLy81sonktP5rX8uOB6PrY7A62HjrNksTjLNx2lAPp\nOSxKOMaihGME+XlyX7u6DIyqT/0A7+uWUUSkIlMBICLiglLSspmych9fbzxE3h8373pYzNwRXpt+\nbevQsXEA7uXkk3U3s4k29WvQpn4NXugVSuKRDL7ZfJh5mw+TmpnHpF+S+HBFEr3CghnerTHtGtQw\nOrKISLmmAkBExIVsO3SGj1YksSjhKHZH4bHQWn4M6lCPvm3qUs3b3diAl2EymWhVpxqt6lRjdK8W\nLNtxnJnrD7Byz8miqwLtGtTg0Rsb07NlLcxaSUhE5AIqAEREXEDikTP8a8kulp+30k730Jo8dmMT\nOjb2x2SqeANlD4uZ3uG16R1em93HM/lk5T6+2XyETSmneOzzTYSFVOX5W0PpHlqzQr4+EZFrRQWA\niEgllnwym3//tJtv448AhdNp7o4I4dEbG9OydlWD05Wd5rX8eOe+CJ6/NZTpq5OJW51M4pEMYqdv\noF2DGvzt1uZ0bhJodEwRkXJBBYCISCV0JqeA/yzdzYy1KVj/mOtzV0QIz/VsTqNAH4PTXTtBVb0Y\n3asFw7o24uNf9xG3OplNKad4YMo6/tKyFv+4syUNAirv6xcRKQ0VACIilYjd7uCrjQd5Z8muoqU8\nu4fW5PlbQ2lVp5rB6a6fAF9PXry9JcO6NuKDn/cyc/0Blu44zq+7TzC8WyOe7NEUH0/9CRQR16R3\nPxGRSiL+4GleXpBA/KEzADQL8uXVu8Po3NR1p77UqurFa/e0YkinBvzz++2s3HOSSb8kMff3Q4y9\nK4zerYJ1f4CIuBwVACIiFdzZfBvv/biLqav243CAn6eFZ3s2Z0inBuVmKU+jNavlx2dDO/DT9uO8\n/sMODqTn8MQXv3PrDbV47Z5W1KrqZXREEZHrRgWAiEgFtnZfGi/M3UpKWg4A90SG8OIdLQny04D2\nz0wmE7eGBXNj85pMWr6XSb8k8eP246xJSuP/bm/JwKh6WjZURFyCCgARkQooJ9/KWwt38vnaFABq\nV/Pizb7h9GgRZHCy8s/L3Y3nbg3l9ta1eWHOVuIPneHF+dv4YdsR3usfSXA1FU8iUrnp2rCISAWT\ncPgMd074rWjw/0B0fX78640a/DupRXBV5j3Rhb/f0RIvdzOr9qZx2/u/snDbUaOjiYhcUyoAREQq\nCLvdwUcrkug7aRX7TmQTXNWLL4ZH82bfcPy8yvcOvuWVm9nE8G6N+eHpboTXqcaZswU88cXv/O2r\neDJzC4yOJyJyTagAEBGpAFIzcnno03WMW7STApuDXmHBLHqmG11ceIWfstSkpi9zR3TmqR5NMZtg\n7u+HuHPCbyQeOWN0NBGRMqcCQESknFu3L407JvzGqr1pVHF3Y1y/cD4c3JYaPh5GR6tUPCxmnr8t\nlC8f60Sd6lVIScuh76TVfLnhgNHRRETKlAoAEZFyyuFwMOXXfTzwyTpOZOYRWsuP75/uysAO9bV2\n/TUU1dCfH57uys0tgsi32nlh7jZGfR3P2Xyb0dFERMqECgARkXIoM7dwLvobC3dgszu4JzKE+U92\npklNX6OjuYTq3h58MqQ9o24LxWyCrzcdou+kVew/mW10NBGRq6YCQESknEk6kUWfD1axKOEY7m4m\nXusTxn8GROLtoZWbryez2cSTPZoyY3g0gb4e7DyWSZ8PfmPlnhNGRxMRuSoqAEREypGVe05wz8RV\n7DuZTe1qXnz5WCce6tRQU34M1LlJID883Y229auTkWslZtoG4lYn43A4jI4mInJFVACIiJQTn69J\nJmbaBjJzrbRrUINvn+pK2/o1jI4lQK2qXsx6tCP3tq2Lze5g7LeJvDg/gXyr3ehoIiJOUwEgImIw\nq83OywsS+MeCRGx2B/3a1OGL4dHU9PM0Opqcx9Pixr/6t+bF21tgMsGs9Qd4aOo60rPzjY4mIuIU\nFQAiIgbKzC0gdvoGPltTuKvv6F6hvHd/BF7ubgYnk5KYTCYevbEJUx9uj6+nhXX707n3w9WkpOnm\nYBGpOFQAiIgYJDUjlwEfr2XlnpNUcXfjo8HteKJ7U833rwBublGLeU90pk71Kuw/mc29H64m/uBp\no2OJiJSKCgAREQMkncii34er2X40g0BfD758rCO9WgUbHUuc0LyWH/Of6ExYSFVOZuUzcPJalu9M\nNTqWiMhlOV0AZGVl8eyzzxISEoKXlxeRkZHMnj27VG1TU1OJiYkhMDAQb29vOnXqxLJly4qdk5GR\nwRtvvEH37t0JDg7G19eX8PBw3n77bXJzcy94zIKCAl599VUaNmyIp6cnLVq0YMKECc6+LBGR62ZT\nyinu/XA1h06dpWGAN3NHdKZ13epGx5IrEFS1cKWmbs0COVtgY/hnG7VzsIiUe04XAP369SMuLo6x\nY8eyaNEioqKiGDRoEDNnzrxku7y8PG655RaWLVvG+PHjWbBgAbVq1aJXr16sWLGi6LwDBw7w/vvv\n07ZtWyZPnsy3337LfffdxyuvvMKdd955wbJrTzzxBG+99RZPPvkkS5YsoW/fvjzzzDO8+eabzr40\nEZFrbun24zz4yVpO5xQQUbcac0Z0pkGAj9Gx5Cr4elr4NCaqaIWgF+Zu44Of92iZUBEpt5zaVWbh\nwoX89NNPzJw5k0GDBgHQo0cPUlJSGDVqFAMGDMDNreQb16ZOnUpCQgKrV6+mU6dORW0jIiIYPXo0\n69atA6BRo0YkJyfj4/O/P4g333wzPj4+jBo1ilWrVtG1a1cAEhMTmTp1Km+88QajRo0CoHv37qSl\npfH666/z+OOP4+/v7+SPRETk2pi/+RDPf70Vm91B99CaTHygLT6e2tyrMnB3M/Ov/q2pXc2LD5bv\n5V8/7iYz18qY3i10T4eIlDtOXQGYP38+vr6+9O/fv9jx2NhYjhw5UjSIv1jb0NDQosE/gMViYfDg\nwaxfv57Dhw8D4OPjU2zwf06HDh0AOHjwYNGxb775BofDQWxs7AV5zp49y+LFi515eSIi18zMdQd4\n7qv4omU+pwxpr8F/JWMymXj+tlD+fkdLAD7+dR8vfZOAza4rASJSvjhVACQkJNCyZUssluJ/tFq3\nbl30/Uu1PXdeSW0TExMv+dw///wzAGFhYcUes2bNmgQHF79xrjR5RESulym/7uPF+dtwOOChjg34\nV/8I3N20BkNlNbxbY8b1C8dkOlf4baHApg3DRKT8cOrjp7S0NBo3bnzB8XPTbNLS0i7ZtqTpOKVp\nu3XrVt555x369u1brIi42GP6+Pjg4eFxycc8X2JiIna73pyvVEFBQdHX+Ph4g9OI0dQf/sfhcPBl\nQiazEjIA6NfSj3sbWtm2bavBya4vV+wTLT3hb538+c+adBZsOcKRE+mM7hKAh5umA7lif5BLU5+4\nemazmfr165f6fKevP19qLuPl5jleSdvk5GTuvPNO6tWrxyeffFKmec6xWq3YbLZSnLine truncated
+      "image/png": "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 truncated
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x26fb1488748>"
+       "<matplotlib.figure.Figure at 0x2c214106748>"
       ]
      },
      "metadata": {},
@@ -988,9 +986,9 @@
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAvYAAAEwCAYAAADcuDUUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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 truncated
+      "image/png": "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 truncated
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x26fb4895630>"
+       "<matplotlib.figure.Figure at 0x2c2142b5da0>"
       ]
      },
      "metadata": {},
@@ -1111,9 +1109,9 @@
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAusAAAEaCAYAAACo3xkcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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 truncated
+      "image/png": "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 truncated
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x26fb4982c50>"
+       "<matplotlib.figure.Figure at 0x2c2143c44e0>"
       ]
      },
      "metadata": {},
@@ -1164,9 +1162,9 @@
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAswAAAEbCAYAAADOAtIQAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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 truncated
+      "image/png": "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 truncated
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x26fb4a60cf8>"
+       "<matplotlib.figure.Figure at 0x2c214558e48>"
       ]
      },
      "metadata": {},
@@ -1195,12 +1193,27 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "ca45591540ed4bdbb79da6bcbdecb7aa",
+       "model_id": "9fd4281843f941b8a16ca99c17362120",
        "version_major": 2,
        "version_minor": 0
       },
+      "text/html": [
+       "<p>Failed to display Jupyter Widget of type <code>interactive</code>.</p>\n",
+       "<p>\n",
+       "  If you're reading this message in the Jupyter Notebook or JupyterLab Notebook, it may mean\n",
+       "  that the widgets JavaScript is still loading. If this message persists, it\n",
+       "  likely means that the widgets JavaScript library is either not installed or\n",
+       "  not enabled. See the <a href=\"https://ipywidgets.readthedocs.io/en/stable/user_install.html\">Jupyter\n",
+       "  Widgets Documentation</a> for setup instructions.\n",
+       "</p>\n",
+       "<p>\n",
+       "  If you're reading this message in another frontend (for example, a static\n",
+       "  rendering on GitHub or <a href=\"https://nbviewer.jupyter.org/\">NBViewer</a>),\n",
+       "  it may mean that your frontend doesn't currently support widgets.\n",
+       "</p>\n"
+      ],
       "text/plain": [
-       "A Jupyter Widget"
+       "interactive(children=(FloatSlider(value=5.0, description='mu', max=10.0), FloatSlider(value=0.6000000000000001, description='variance', max=1.0, min=0.2), Output()), _dom_classes=('widget-interact',))"
       ]
      },
      "metadata": {},
@@ -1395,8 +1408,8 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "0.131146572034\n",
-      "0.131146572034\n"
+      "0.13114657203397997\n",
+      "0.13114657203397997\n"
      ]
     }
    ],
@@ -1452,9 +1465,9 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "[-1.615  5.798 -1.458 -2.189  4.282  0.664  5.282\n",
-      "  0.212 -1.687  5.052 -2.256  2.516  0.403  0.966\n",
-      "  4.009]\n"
+      "[ 1.77   3.378 -3.014  5.337  4.285  4.797  1.477\n",
+      "  1.351  5.739  4.075 -1.575  0.681  0.202 -0.747\n",
+      "  1.392]\n"
      ]
     }
    ],
@@ -1540,9 +1553,9 @@
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAxMAAAEaCAYAAABqygqxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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 truncated
+      "image/png": "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 truncated
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x26fb4995518>"
+       "<matplotlib.figure.Figure at 0x2c2144ed978>"
       ]
      },
      "metadata": {},
@@ -1571,9 +1584,7 @@
   {
    "cell_type": "code",
    "execution_count": 29,
-   "metadata": {
-    "collapsed": true
-   },
+   "metadata": {},
    "outputs": [],
    "source": [
     "from numpy.random import randn\n",
@@ -1595,9 +1606,9 @@
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAvUAAAEaCAYAAACRuKhHAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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 truncated
+      "image/png": "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 truncated
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x26fb3210978>"
+       "<matplotlib.figure.Figure at 0x2c214332908>"
       ]
      },
      "metadata": {},
@@ -1621,9 +1632,7 @@
   {
    "cell_type": "code",
    "execution_count": 31,
-   "metadata": {
-    "collapsed": true
-   },
+   "metadata": {},
    "outputs": [],
    "source": [
     "import random\n",
@@ -1646,9 +1655,9 @@
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAukAAAEdCAYAAACxL/mZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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 truncated
+      "image/png": "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 truncated
       "text/plain": [
-       "<matplotlib.figure.Figure at 0x26fb4967400>"
+       "<matplotlib.figure.Figure at 0x2c214169b38>"
       ]
      },
      "metadata": {},
@@ -1679,7 +1688,63 @@
     "from filterpy.stats import rand_student_t\n",
     "```\n",
     "\n",
-    "While I'll not cover it here, statistics has defined ways of describing the shape of a probability distribution by how it varies from an exponential distribution. The normal distribution is shaped symmetrically around the mean - like a bell curve. However, a probability distribution can be asymmetrical around the mean. The measure of this is called [*skew*](https://en.wikipedia.org/wiki/Skewness). The tails can be shortened, fatter, thinner, or otherwise shaped differently from an exponential distribution. The measure of this is called [*kurtosis*](https://en.wikipedia.org/wiki/Kurtosis)."
+    "While I'll not cover it here, statistics has defined ways of describing the shape of a probability distribution by how it varies from an exponential distribution. The normal distribution is shaped symmetrically around the mean - like a bell curve. However, a probability distribution can be asymmetrical around the mean. The measure of this is called [*skew*](https://en.wikipedia.org/wiki/Skewness). The tails can be shortened, fatter, thinner, or otherwise shaped differently from an exponential distribution. The measure of this is called [*kurtosis*](https://en.wikipedia.org/wiki/Kurtosis). the `scipy.stats` module contains the function `describe` which computes these statistics, among others."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 33,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "DescribeResult(nobs=5000, minmax=(1.3813891957392315, 19.42199152705908), mean=10.007529817656724, variance=2.7772693671119986, skewness=-0.07073510230931251, kurtosis=1.4943039523916015)"
+      ]
+     },
+     "execution_count": 33,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import scipy\n",
+    "scipy.stats.describe(zs)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Let's examine two normal populations, one small, one large:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 34,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "DescribeResult(nobs=10, minmax=(-0.8175084513413164, 1.1865956580866834), mean=0.2940337215902996, variance=0.6230354047551284, skewness=-0.41279400723365806, kurtosis=-1.3123468690692666)\n",
+      "\n",
+      "DescribeResult(nobs=300000, minmax=(-4.7480566375065525, 4.550774022502299), mean=0.0005748027839955969, variance=1.0005892638782288, skewness=6.280909739979528e-05, kurtosis=0.006885785867941863)\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(scipy.stats.describe(np.random.randn(10)))\n",
+    "print()\n",
+    "print(scipy.stats.describe(np.random.randn(300000)))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The small sample has very non-zero skew and kurtosis because the small number of samples is not well distributed around the mean of 0. You can see this also by comparing the computed mean and variance with the theoretical mean of 0 and variance 1. In comparison the large sample's mean and variance are very close to the theoretical values, and both the skew and kurtosis are near zero."
    ]
   },
   {