{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Lesson 26: Hacker stats I\n", "\n", "
" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [ { "data": { "text/html": [ " \n", "
\n", " \n", " Loading BokehJS ...\n", "
\n" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/javascript": [ "'use strict';\n", "(function(root) {\n", " function now() {\n", " return new Date();\n", " }\n", "\n", " const force = true;\n", "\n", " if (typeof root._bokeh_onload_callbacks === \"undefined\" || force === true) {\n", " root._bokeh_onload_callbacks = [];\n", " root._bokeh_is_loading = undefined;\n", " }\n", "\n", "const JS_MIME_TYPE = 'application/javascript';\n", " const HTML_MIME_TYPE = 'text/html';\n", " const EXEC_MIME_TYPE = 'application/vnd.bokehjs_exec.v0+json';\n", " const CLASS_NAME = 'output_bokeh rendered_html';\n", "\n", " /**\n", " * Render data to the DOM node\n", " */\n", " function render(props, node) {\n", " const script = document.createElement(\"script\");\n", " node.appendChild(script);\n", " }\n", "\n", " /**\n", " * Handle when an output is cleared or removed\n", " */\n", " function handleClearOutput(event, handle) {\n", " function drop(id) {\n", " const view = Bokeh.index.get_by_id(id)\n", " if (view != null) {\n", " view.model.document.clear()\n", " Bokeh.index.delete(view)\n", " }\n", " }\n", "\n", " const cell = handle.cell;\n", "\n", " const id = cell.output_area._bokeh_element_id;\n", " const server_id = cell.output_area._bokeh_server_id;\n", "\n", " // Clean up Bokeh references\n", " if (id != null) {\n", " drop(id)\n", " }\n", "\n", " if (server_id !== undefined) {\n", " // Clean up Bokeh references\n", " const cmd_clean = \"from bokeh.io.state import curstate; print(curstate().uuid_to_server['\" + server_id + \"'].get_sessions()[0].document.roots[0]._id)\";\n", " cell.notebook.kernel.execute(cmd_clean, {\n", " iopub: {\n", " output: function(msg) {\n", " const id = msg.content.text.trim()\n", " drop(id)\n", " }\n", " }\n", " });\n", " // Destroy server and session\n", " const cmd_destroy = \"import bokeh.io.notebook as ion; ion.destroy_server('\" + server_id + \"')\";\n", " cell.notebook.kernel.execute(cmd_destroy);\n", " }\n", " }\n", "\n", " /**\n", " * Handle when a new output is added\n", " */\n", " function handleAddOutput(event, handle) {\n", " const output_area = handle.output_area;\n", " const output = handle.output;\n", "\n", " // limit handleAddOutput to display_data with EXEC_MIME_TYPE content only\n", " if ((output.output_type != \"display_data\") || (!Object.prototype.hasOwnProperty.call(output.data, EXEC_MIME_TYPE))) {\n", " return\n", " }\n", "\n", " const toinsert = output_area.element.find(\".\" + CLASS_NAME.split(' ')[0]);\n", "\n", " if (output.metadata[EXEC_MIME_TYPE][\"id\"] !== undefined) {\n", " toinsert[toinsert.length - 1].firstChild.textContent = output.data[JS_MIME_TYPE];\n", " // store reference to embed id on output_area\n", " output_area._bokeh_element_id = output.metadata[EXEC_MIME_TYPE][\"id\"];\n", " }\n", " if (output.metadata[EXEC_MIME_TYPE][\"server_id\"] !== undefined) {\n", " const bk_div = document.createElement(\"div\");\n", " bk_div.innerHTML = output.data[HTML_MIME_TYPE];\n", " const script_attrs = bk_div.children[0].attributes;\n", " for (let i = 0; i < script_attrs.length; i++) {\n", " toinsert[toinsert.length - 1].firstChild.setAttribute(script_attrs[i].name, script_attrs[i].value);\n", " toinsert[toinsert.length - 1].firstChild.textContent = bk_div.children[0].textContent\n", " }\n", " // store reference to server id on output_area\n", " output_area._bokeh_server_id = output.metadata[EXEC_MIME_TYPE][\"server_id\"];\n", " }\n", " }\n", "\n", " function register_renderer(events, OutputArea) {\n", "\n", " function append_mime(data, metadata, element) {\n", " // create a DOM node to render to\n", " const toinsert = this.create_output_subarea(\n", " metadata,\n", " CLASS_NAME,\n", " EXEC_MIME_TYPE\n", " );\n", " this.keyboard_manager.register_events(toinsert);\n", " // Render to node\n", " const props = {data: data, metadata: metadata[EXEC_MIME_TYPE]};\n", " render(props, toinsert[toinsert.length - 1]);\n", " element.append(toinsert);\n", " return toinsert\n", " }\n", "\n", " /* Handle when an output is cleared or removed */\n", " events.on('clear_output.CodeCell', handleClearOutput);\n", " events.on('delete.Cell', handleClearOutput);\n", "\n", " /* Handle when a new output is added */\n", " events.on('output_added.OutputArea', handleAddOutput);\n", "\n", " /**\n", " * Register the mime type and append_mime function with output_area\n", " */\n", " OutputArea.prototype.register_mime_type(EXEC_MIME_TYPE, append_mime, {\n", " /* Is output safe? */\n", " safe: true,\n", " /* Index of renderer in `output_area.display_order` */\n", " index: 0\n", " });\n", " }\n", "\n", " // register the mime type if in Jupyter Notebook environment and previously unregistered\n", " if (root.Jupyter !== undefined) {\n", " const events = require('base/js/events');\n", " const OutputArea = require('notebook/js/outputarea').OutputArea;\n", "\n", " if (OutputArea.prototype.mime_types().indexOf(EXEC_MIME_TYPE) == -1) {\n", " register_renderer(events, OutputArea);\n", " }\n", " }\n", " if (typeof (root._bokeh_timeout) === \"undefined\" || force === true) {\n", " root._bokeh_timeout = Date.now() + 5000;\n", " root._bokeh_failed_load = false;\n", " }\n", "\n", " const NB_LOAD_WARNING = {'data': {'text/html':\n", " \"
\\n\"+\n", " \"

\\n\"+\n", " \"BokehJS does not appear to have successfully loaded. If loading BokehJS from CDN, this \\n\"+\n", " \"may be due to a slow or bad network connection. Possible fixes:\\n\"+\n", " \"

\\n\"+\n", " \"\\n\"+\n", " \"\\n\"+\n", " \"from bokeh.resources import INLINE\\n\"+\n", " \"output_notebook(resources=INLINE)\\n\"+\n", " \"\\n\"+\n", " \"
\"}};\n", "\n", " function display_loaded(error = null) {\n", " const el = document.getElementById(\"c06fdd8d-f8fb-433a-bff4-51416dcb717a\");\n", " if (el != null) {\n", " const html = (() => {\n", " if (typeof root.Bokeh === \"undefined\") {\n", " if (error == null) {\n", " return \"BokehJS is loading ...\";\n", " } else {\n", " return \"BokehJS failed to load.\";\n", " }\n", " } else {\n", " const prefix = `BokehJS ${root.Bokeh.version}`;\n", " if (error == null) {\n", " return `${prefix} successfully loaded.`;\n", " } else {\n", " return `${prefix} encountered errors while loading and may not function as expected.`;\n", " }\n", " }\n", " })();\n", " el.innerHTML = html;\n", "\n", " if (error != null) {\n", " const wrapper = document.createElement(\"div\");\n", " wrapper.style.overflow = \"auto\";\n", " wrapper.style.height = \"5em\";\n", " wrapper.style.resize = \"vertical\";\n", " const content = document.createElement(\"div\");\n", " content.style.fontFamily = \"monospace\";\n", " content.style.whiteSpace = \"pre-wrap\";\n", " content.style.backgroundColor = \"rgb(255, 221, 221)\";\n", " content.textContent = error.stack ?? error.toString();\n", " wrapper.append(content);\n", " el.append(wrapper);\n", " }\n", " } else if (Date.now() < root._bokeh_timeout) {\n", " setTimeout(() => display_loaded(error), 100);\n", " }\n", " }\n", "\n", " function run_callbacks() {\n", " try {\n", " root._bokeh_onload_callbacks.forEach(function(callback) {\n", " if (callback != null)\n", " callback();\n", " });\n", " } finally {\n", " delete root._bokeh_onload_callbacks\n", " }\n", " console.debug(\"Bokeh: all callbacks have finished\");\n", " }\n", "\n", " function load_libs(css_urls, js_urls, callback) {\n", " if (css_urls == null) css_urls = [];\n", " if (js_urls == null) js_urls = [];\n", "\n", " root._bokeh_onload_callbacks.push(callback);\n", " if (root._bokeh_is_loading > 0) {\n", " console.debug(\"Bokeh: BokehJS is being loaded, scheduling callback at\", now());\n", " return null;\n", " }\n", " if (js_urls == null || js_urls.length === 0) {\n", " run_callbacks();\n", " return null;\n", " }\n", " console.debug(\"Bokeh: BokehJS not loaded, scheduling load and callback at\", now());\n", " root._bokeh_is_loading = css_urls.length + js_urls.length;\n", "\n", " function on_load() {\n", " root._bokeh_is_loading--;\n", " if (root._bokeh_is_loading === 0) {\n", " console.debug(\"Bokeh: all BokehJS libraries/stylesheets loaded\");\n", " run_callbacks()\n", " }\n", " }\n", "\n", " function on_error(url) {\n", " console.error(\"failed to load \" + url);\n", " }\n", "\n", " for (let i = 0; i < css_urls.length; i++) {\n", " const url = css_urls[i];\n", " const element = document.createElement(\"link\");\n", " element.onload = on_load;\n", " element.onerror = on_error.bind(null, url);\n", " element.rel = \"stylesheet\";\n", " element.type = \"text/css\";\n", " element.href = url;\n", " console.debug(\"Bokeh: injecting link tag for BokehJS stylesheet: \", url);\n", " document.body.appendChild(element);\n", " }\n", "\n", " for (let i = 0; i < js_urls.length; i++) {\n", " const url = js_urls[i];\n", " const element = document.createElement('script');\n", " element.onload = on_load;\n", " element.onerror = on_error.bind(null, url);\n", " element.async = false;\n", " element.src = url;\n", " console.debug(\"Bokeh: injecting script tag for BokehJS library: \", url);\n", " document.head.appendChild(element);\n", " }\n", " };\n", "\n", " function inject_raw_css(css) {\n", " const element = document.createElement(\"style\");\n", " element.appendChild(document.createTextNode(css));\n", " document.body.appendChild(element);\n", " }\n", "\n", " const js_urls = [\"https://cdn.bokeh.org/bokeh/release/bokeh-3.4.0.min.js\", \"https://cdn.bokeh.org/bokeh/release/bokeh-gl-3.4.0.min.js\", \"https://cdn.bokeh.org/bokeh/release/bokeh-widgets-3.4.0.min.js\", \"https://cdn.bokeh.org/bokeh/release/bokeh-tables-3.4.0.min.js\", \"https://cdn.bokeh.org/bokeh/release/bokeh-mathjax-3.4.0.min.js\"];\n", " const css_urls = [];\n", "\n", " const inline_js = [ function(Bokeh) {\n", " Bokeh.set_log_level(\"info\");\n", " },\n", "function(Bokeh) {\n", " }\n", " ];\n", "\n", " function run_inline_js() {\n", " if (root.Bokeh !== undefined || force === true) {\n", " try {\n", " for (let i = 0; i < inline_js.length; i++) {\n", " inline_js[i].call(root, root.Bokeh);\n", " }\n", "\n", " } catch (error) {display_loaded(error);throw error;\n", " }if (force === true) {\n", " display_loaded();\n", " }} else if (Date.now() < root._bokeh_timeout) {\n", " setTimeout(run_inline_js, 100);\n", " } else if (!root._bokeh_failed_load) {\n", " console.log(\"Bokeh: BokehJS failed to load within specified timeout.\");\n", " root._bokeh_failed_load = true;\n", " } else if (force !== true) {\n", " const cell = $(document.getElementById(\"c06fdd8d-f8fb-433a-bff4-51416dcb717a\")).parents('.cell').data().cell;\n", " cell.output_area.append_execute_result(NB_LOAD_WARNING)\n", " }\n", " }\n", "\n", " if (root._bokeh_is_loading === 0) {\n", " console.debug(\"Bokeh: BokehJS loaded, going straight to plotting\");\n", " run_inline_js();\n", " } else {\n", " load_libs(css_urls, js_urls, function() {\n", " console.debug(\"Bokeh: BokehJS plotting callback run at\", now());\n", " run_inline_js();\n", " });\n", " }\n", "}(window));" ], "application/vnd.bokehjs_load.v0+json": "'use strict';\n(function(root) {\n function now() {\n return new Date();\n }\n\n const force = true;\n\n if (typeof root._bokeh_onload_callbacks === \"undefined\" || force === true) {\n root._bokeh_onload_callbacks = [];\n root._bokeh_is_loading = undefined;\n }\n\n\n if (typeof (root._bokeh_timeout) === \"undefined\" || force === true) {\n root._bokeh_timeout = Date.now() + 5000;\n root._bokeh_failed_load = false;\n }\n\n const NB_LOAD_WARNING = {'data': {'text/html':\n \"
\\n\"+\n \"

\\n\"+\n \"BokehJS does not appear to have successfully loaded. If loading BokehJS from CDN, this \\n\"+\n \"may be due to a slow or bad network connection. Possible fixes:\\n\"+\n \"

\\n\"+\n \"\\n\"+\n \"\\n\"+\n \"from bokeh.resources import INLINE\\n\"+\n \"output_notebook(resources=INLINE)\\n\"+\n \"\\n\"+\n \"
\"}};\n\n function display_loaded(error = null) {\n const el = document.getElementById(\"c06fdd8d-f8fb-433a-bff4-51416dcb717a\");\n if (el != null) {\n const html = (() => {\n if (typeof root.Bokeh === \"undefined\") {\n if (error == null) {\n return \"BokehJS is loading ...\";\n } else {\n return \"BokehJS failed to load.\";\n }\n } else {\n const prefix = `BokehJS ${root.Bokeh.version}`;\n if (error == null) {\n return `${prefix} successfully loaded.`;\n } else {\n return `${prefix} encountered errors while loading and may not function as expected.`;\n }\n }\n })();\n el.innerHTML = html;\n\n if (error != null) {\n const wrapper = document.createElement(\"div\");\n wrapper.style.overflow = \"auto\";\n wrapper.style.height = \"5em\";\n wrapper.style.resize = \"vertical\";\n const content = document.createElement(\"div\");\n content.style.fontFamily = \"monospace\";\n content.style.whiteSpace = \"pre-wrap\";\n content.style.backgroundColor = \"rgb(255, 221, 221)\";\n content.textContent = error.stack ?? error.toString();\n wrapper.append(content);\n el.append(wrapper);\n }\n } else if (Date.now() < root._bokeh_timeout) {\n setTimeout(() => display_loaded(error), 100);\n }\n }\n\n function run_callbacks() {\n try {\n root._bokeh_onload_callbacks.forEach(function(callback) {\n if (callback != null)\n callback();\n });\n } finally {\n delete root._bokeh_onload_callbacks\n }\n console.debug(\"Bokeh: all callbacks have finished\");\n }\n\n function load_libs(css_urls, js_urls, callback) {\n if (css_urls == null) css_urls = [];\n if (js_urls == null) js_urls = [];\n\n root._bokeh_onload_callbacks.push(callback);\n if (root._bokeh_is_loading > 0) {\n console.debug(\"Bokeh: BokehJS is being loaded, scheduling callback at\", now());\n return null;\n }\n if (js_urls == null || js_urls.length === 0) {\n run_callbacks();\n return null;\n }\n console.debug(\"Bokeh: BokehJS not loaded, scheduling load and callback at\", now());\n root._bokeh_is_loading = css_urls.length + js_urls.length;\n\n function on_load() {\n root._bokeh_is_loading--;\n if (root._bokeh_is_loading === 0) {\n console.debug(\"Bokeh: all BokehJS libraries/stylesheets loaded\");\n run_callbacks()\n }\n }\n\n function on_error(url) {\n console.error(\"failed to load \" + url);\n }\n\n for (let i = 0; i < css_urls.length; i++) {\n const url = css_urls[i];\n const element = document.createElement(\"link\");\n element.onload = on_load;\n element.onerror = on_error.bind(null, url);\n element.rel = \"stylesheet\";\n element.type = \"text/css\";\n element.href = url;\n console.debug(\"Bokeh: injecting link tag for BokehJS stylesheet: \", url);\n document.body.appendChild(element);\n }\n\n for (let i = 0; i < js_urls.length; i++) {\n const url = js_urls[i];\n const element = document.createElement('script');\n element.onload = on_load;\n element.onerror = on_error.bind(null, url);\n element.async = false;\n element.src = url;\n console.debug(\"Bokeh: injecting script tag for BokehJS library: \", url);\n document.head.appendChild(element);\n }\n };\n\n function inject_raw_css(css) {\n const element = document.createElement(\"style\");\n element.appendChild(document.createTextNode(css));\n document.body.appendChild(element);\n }\n\n const js_urls = [\"https://cdn.bokeh.org/bokeh/release/bokeh-3.4.0.min.js\", \"https://cdn.bokeh.org/bokeh/release/bokeh-gl-3.4.0.min.js\", \"https://cdn.bokeh.org/bokeh/release/bokeh-widgets-3.4.0.min.js\", \"https://cdn.bokeh.org/bokeh/release/bokeh-tables-3.4.0.min.js\", \"https://cdn.bokeh.org/bokeh/release/bokeh-mathjax-3.4.0.min.js\"];\n const css_urls = [];\n\n const inline_js = [ function(Bokeh) {\n Bokeh.set_log_level(\"info\");\n },\nfunction(Bokeh) {\n }\n ];\n\n function run_inline_js() {\n if (root.Bokeh !== undefined || force === true) {\n try {\n for (let i = 0; i < inline_js.length; i++) {\n inline_js[i].call(root, root.Bokeh);\n }\n\n } catch (error) {display_loaded(error);throw error;\n }if (force === true) {\n display_loaded();\n }} else if (Date.now() < root._bokeh_timeout) {\n setTimeout(run_inline_js, 100);\n } else if (!root._bokeh_failed_load) {\n console.log(\"Bokeh: BokehJS failed to load within specified timeout.\");\n root._bokeh_failed_load = true;\n } else if (force !== true) {\n const cell = $(document.getElementById(\"c06fdd8d-f8fb-433a-bff4-51416dcb717a\")).parents('.cell').data().cell;\n cell.output_area.append_execute_result(NB_LOAD_WARNING)\n }\n }\n\n if (root._bokeh_is_loading === 0) {\n console.debug(\"Bokeh: BokehJS loaded, going straight to plotting\");\n run_inline_js();\n } else {\n load_libs(css_urls, js_urls, function() {\n console.debug(\"Bokeh: BokehJS plotting callback run at\", now());\n run_inline_js();\n });\n }\n}(window));" }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import numpy as np\n", "import pandas as pd\n", "\n", "import iqplot\n", "\n", "import bokeh.io\n", "import bokeh.plotting\n", "\n", "bokeh.io.output_notebook()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
\n", "\n", "When the field of statistics was in its early days, the practitioners did not have computers. They were therefore left to use pen and paper to compute things like confidence intervals. Despite their toils, you will soon see that with just a little bit of programming experience, you can perform lots of the statistical analyses that may seem baffling when done with pen and paper.\n", "\n", "At the heart of this \"hacker statistics\" is the ability to draw random numbers. We will focus on **bootstrap** methods in particular.\n", "\n", "To motivate this study, we will work with data measured by Peter and Rosemary Grant on the island of Daphne Major on the Galápagos. They have been going to the island every year for over forty years and have been taking a careful inventory of the finches there. We will look at the finch *Geospiza scandens*. The Grants measured the depths of the beaks (defined as the top-to-bottom thickness of the beak) of all of the finches of this species on the island. We will consider their measurements from 1975 and from 2012. We will investigate how the beaks got deeper over time.\n", "\n", "The data are from the book Grants' book [*40 years of evolution: Darwin's finches on Daphne Major Island*](http://www.worldcat.org/oclc/854285415). They were generous and made their data publicly available on the [Dryad data repository](http://dx.doi.org/10.5061/dryad.g6g3h). In general, it is a very good idea to put your published data in public data repositories, both to preserve the data and also to make your findings public.\n", "\n", "Ok, let's start by loading in the data. You converted the Grants' data into a single data frame in the exercises. Let's load the data, which are available in the file `~/git/bootcamp/data/grant_complete.csv`." ] }, { "cell_type": "code", "execution_count": 2, "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", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
bandbeak depth (mm)beak length (mm)speciesyear
0201238.059.25fortis1973
12012610.4511.35fortis1973
2201289.5510.15fortis1973
3201298.759.95fortis1973
42013310.1511.55fortis1973
\n", "
" ], "text/plain": [ " band beak depth (mm) beak length (mm) species year\n", "0 20123 8.05 9.25 fortis 1973\n", "1 20126 10.45 11.35 fortis 1973\n", "2 20128 9.55 10.15 fortis 1973\n", "3 20129 8.75 9.95 fortis 1973\n", "4 20133 10.15 11.55 fortis 1973" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df = pd.read_csv('data/grant_complete.csv', comment='#')\n", "\n", "df.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's trim down the data frame to only include *G. scandens* from 1975 and 2012 and only include the columns we need." ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "df = df.loc[\n", " (df[\"species\"] == \"scandens\") & (df[\"year\"].isin([1975, 2012])),\n", " [\"year\", \"beak depth (mm)\"],\n", "]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's take a look at the ECDFs for these two years." ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "
\n" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/javascript": [ "(function(root) {\n", " function embed_document(root) {\n", " const docs_json = {\"6c137f60-05ae-4ef2-a7a9-00837278ad35\":{\"version\":\"3.4.0\",\"title\":\"Bokeh Application\",\"roots\":[{\"type\":\"object\",\"name\":\"Figure\",\"id\":\"p1001\",\"attributes\":{\"x_range\":{\"type\":\"object\",\"name\":\"DataRange1d\",\"id\":\"p1002\"},\"y_range\":{\"type\":\"object\",\"name\":\"DataRange1d\",\"id\":\"p1003\"},\"x_scale\":{\"type\":\"object\",\"name\":\"LinearScale\",\"id\":\"p1010\"},\"y_scale\":{\"type\":\"object\",\"name\":\"LinearScale\",\"id\":\"p1011\"},\"title\":{\"type\":\"object\",\"name\":\"Title\",\"id\":\"p1008\"},\"renderers\":[{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1039\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1033\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1034\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1035\"},\"data\":{\"type\":\"map\",\"entries\":[[\"x\",{\"type\":\"ndarray\",\"array\":{\"type\":\"bytes\",\"data\":\"mpmZmZmZH0CamZmZmZkfQAAAAAAAACBAAAAAAAAAIEAAAAAAAAAgQAAAAAAAACBAAAAAAAAAIEAAAAAAAAAgQDMzMzMzMyBAMzMzMzMzIEBmZmZmZmYgQGZmZmZmZiBAZmZmZmZmIEBmZmZmZmYgQGZmZmZmZiBAZmZmZmZmIECamZmZmZkgQJqZmZmZmSBAmpmZmZmZIECamZmZmZkgQJqZmZmZmSBAmpmZmZmZIECamZmZmZkgQJqZmZmZmSBAMzMzMzOzIEAzMzMzM7MgQM3MzMzMzCBAzczMzMzMIEDNzMzMzMwgQM3MzMzMzCBAzczMzMzMIEDNzMzMzMwgQM3MzMzMzCBAzczMzMzMIEAAAAAAAAAhQAAAAAAAACFAAAAAAAAAIUAAAAAAAAAhQAAAAAAAACFAAAAAAAAAIUAAAAAAAAAhQAAAAAAAACFAAAAAAAAAIUAAAAAAAAAhQDMzMzMzMyFAMzMzMzMzIUAzMzMzMzMhQDMzMzMzMyFAMzMzMzMzIUAzMzMzMzMhQDMzMzMzMyFAMzMzMzMzIUAzMzMzMzMhQDMzMzMzMyFAMzMzMzMzIUAzMzMzMzMhQM3MzMzMTCFAzczMzMxMIUBmZmZmZmYhQGZmZmZmZiFAZmZmZmZmIUBmZmZmZmYhQGZmZmZmZiFAZmZmZmZmIUAAAAAAAIAhQAAAAAAAgCFAmpmZmZmZIUCamZmZmZkhQJqZmZmZmSFAmpmZmZmZIUCamZmZmZkhQJqZmZmZmSFAmpmZmZmZIUCamZmZmZkhQJqZmZmZmSFAmpmZmZmZIUDNzMzMzMwhQM3MzMzMzCFAzczMzMzMIUDNzMzMzMwhQM3MzMzMzCFAzczMzMzMIUDNzMzMzMwhQM3MzMzMzCFAzczMzMzMIUDNzMzMzMwhQAAAAAAAACJAAAAAAAAAIkAAAAAAAAAiQAAAAAAAACJAAAAAAAAAIkAAAAAAAAAiQAAAAAAAACJAAAAAAAAAIkAAAAAAAAAiQAAAAAAAACJAAAAAAAAAIkAAAAAAAAAiQBSuR+F6FCJAFK5H4XoUIkCamZmZmRkiQJqZmZmZGSJAmpmZmZkZIkCamZmZmRkiQDMzMzMzMyJAMzMzMzMzIkAzMzMzMzMiQDMzMzMzMyJAMzMzMzMzIkAzMzMzMzMiQDMzMzMzMyJAMzMzMzMzIkAzMzMzMzMiQDMzMzMzMyJAMzMzMzMzIkAzMzMzMzMiQDMzMzMzMyJAMzMzMzMzIkAzMzMzMzMiQDMzMzMzMyJAMzMzMzMzIkAzMzMzMzMiQGZmZmZmZiJAZmZmZmZmIkBmZmZmZmYiQGZmZmZmZiJAZmZmZmZmIkBmZmZmZmYiQGZmZmZmZiJAZmZmZmZmIkCamZmZmZkiQJqZmZmZmSJAmpmZmZmZIkCamZmZmZkiQJqZmZmZmSJAmpmZmZmZIkDhehSuR+EiQOF6FK5H4SJAZmZmZmbmIkBmZmZmZuYiQGZmZmZm5iJAZmZmZmbmIkBmZmZmZuYiQGZmZmZm5iJAZmZmZmbmIkBmZmZmZuYiQDMzMzMzMyNAMzMzMzMzI0AzMzMzMzMjQDMzMzMzMyNAzczMzMxMI0DNzMzMzEwjQGZmZmZmZiNAZmZmZmZmI0BmZmZmZmYjQGZmZmZmZiNAexSuR+F6I0B7FK5H4XojQJqZmZmZmSNAmpmZmZmZI0CamZmZmZkjQJqZmZmZmSNAzczMzMzMI0DNzMzMzMwjQDMzMzMzMyRAMzMzMzMzJEAzMzMzMzMkQDMzMzMzMyRAZmZmZmZmJEBmZmZmZmYkQM3MzMzMzCRAzczMzMzMJEDNzMzMzMwkQM3MzMzMzCRA\"},\"shape\":[174],\"dtype\":\"float64\",\"order\":\"little\"}],[\"y\",{\"type\":\"ndarray\",\"array\":{\"type\":\"bytes\",\"data\":\"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\"},\"shape\":[174],\"dtype\":\"float64\",\"order\":\"little\"}]]}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1040\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1041\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1036\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#1f77b3\",\"line_width\":2}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1037\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#1f77b3\",\"line_alpha\":0.1,\"line_width\":2}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1038\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#1f77b3\",\"line_alpha\":0.2,\"line_width\":2}}}},{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1048\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1042\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1043\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1044\"},\"data\":{\"type\":\"map\"}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1049\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1050\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1045\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":7.9},\"y\":{\"type\":\"value\",\"value\":0},\"angle\":{\"type\":\"value\",\"value\":3.141592653589793},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1046\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":7.9},\"y\":{\"type\":\"value\",\"value\":0},\"angle\":{\"type\":\"value\",\"value\":3.141592653589793},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.1},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1047\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":7.9},\"y\":{\"type\":\"value\",\"value\":0},\"angle\":{\"type\":\"value\",\"value\":3.141592653589793},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.2},\"line_width\":{\"type\":\"value\",\"value\":2}}}}},{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1057\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1051\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1052\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1053\"},\"data\":{\"type\":\"map\"}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1058\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1059\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1054\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":10.4},\"y\":{\"type\":\"value\",\"value\":1},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1055\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":10.4},\"y\":{\"type\":\"value\",\"value\":1},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.1},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1056\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":10.4},\"y\":{\"type\":\"value\",\"value\":1},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.2},\"line_width\":{\"type\":\"value\",\"value\":2}}}}},{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1066\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1060\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1061\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1062\"},\"data\":{\"type\":\"map\",\"entries\":[[\"x\",{\"type\":\"ndarray\",\"array\":{\"type\":\"bytes\",\"data\":\"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\"},\"shape\":[252],\"dtype\":\"float64\",\"order\":\"little\"}],[\"y\",{\"type\":\"ndarray\",\"array\":{\"type\":\"bytes\",\"data\":\"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\"},\"shape\":[252],\"dtype\":\"float64\",\"order\":\"little\"}]]}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1067\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1068\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1063\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#ff7e0e\",\"line_width\":2}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1064\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#ff7e0e\",\"line_alpha\":0.1,\"line_width\":2}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1065\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#ff7e0e\",\"line_alpha\":0.2,\"line_width\":2}}}},{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1075\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1069\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1070\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1071\"},\"data\":{\"type\":\"map\"}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1076\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1077\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1072\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":7.7},\"y\":{\"type\":\"value\",\"value\":0},\"angle\":{\"type\":\"value\",\"value\":3.141592653589793},\"line_color\":{\"type\":\"value\",\"value\":\"#ff7e0e\"},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1073\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":7.7},\"y\":{\"type\":\"value\",\"value\":0},\"angle\":{\"type\":\"value\",\"value\":3.141592653589793},\"line_color\":{\"type\":\"value\",\"value\":\"#ff7e0e\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.1},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1074\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":7.7},\"y\":{\"type\":\"value\",\"value\":0},\"angle\":{\"type\":\"value\",\"value\":3.141592653589793},\"line_color\":{\"type\":\"value\",\"value\":\"#ff7e0e\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.2},\"line_width\":{\"type\":\"value\",\"value\":2}}}}},{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1084\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1078\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1079\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1080\"},\"data\":{\"type\":\"map\"}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1085\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1086\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1081\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":11.0},\"y\":{\"type\":\"value\",\"value\":1},\"line_color\":{\"type\":\"value\",\"value\":\"#ff7e0e\"},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1082\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":11.0},\"y\":{\"type\":\"value\",\"value\":1},\"line_color\":{\"type\":\"value\",\"value\":\"#ff7e0e\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.1},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1083\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":11.0},\"y\":{\"type\":\"value\",\"value\":1},\"line_color\":{\"type\":\"value\",\"value\":\"#ff7e0e\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.2},\"line_width\":{\"type\":\"value\",\"value\":2}}}}}],\"toolbar\":{\"type\":\"object\",\"name\":\"Toolbar\",\"id\":\"p1009\",\"attributes\":{\"tools\":[{\"type\":\"object\",\"name\":\"PanTool\",\"id\":\"p1022\"},{\"type\":\"object\",\"name\":\"WheelZoomTool\",\"id\":\"p1023\",\"attributes\":{\"renderers\":\"auto\"}},{\"type\":\"object\",\"name\":\"BoxZoomTool\",\"id\":\"p1024\",\"attributes\":{\"overlay\":{\"type\":\"object\",\"name\":\"BoxAnnotation\",\"id\":\"p1025\",\"attributes\":{\"syncable\":false,\"level\":\"overlay\",\"visible\":false,\"left\":{\"type\":\"number\",\"value\":\"nan\"},\"right\":{\"type\":\"number\",\"value\":\"nan\"},\"top\":{\"type\":\"number\",\"value\":\"nan\"},\"bottom\":{\"type\":\"number\",\"value\":\"nan\"},\"left_units\":\"canvas\",\"right_units\":\"canvas\",\"top_units\":\"canvas\",\"bottom_units\":\"canvas\",\"line_color\":\"black\",\"line_alpha\":1.0,\"line_width\":2,\"line_dash\":[4,4],\"fill_color\":\"lightgrey\",\"fill_alpha\":0.5}}}},{\"type\":\"object\",\"name\":\"SaveTool\",\"id\":\"p1030\"},{\"type\":\"object\",\"name\":\"ResetTool\",\"id\":\"p1031\"},{\"type\":\"object\",\"name\":\"HelpTool\",\"id\":\"p1032\"}]}},\"toolbar_location\":\"above\",\"left\":[{\"type\":\"object\",\"name\":\"LinearAxis\",\"id\":\"p1017\",\"attributes\":{\"ticker\":{\"type\":\"object\",\"name\":\"BasicTicker\",\"id\":\"p1018\",\"attributes\":{\"mantissas\":[1,2,5]}},\"formatter\":{\"type\":\"object\",\"name\":\"BasicTickFormatter\",\"id\":\"p1019\"},\"axis_label\":\"ECDF\",\"major_label_policy\":{\"type\":\"object\",\"name\":\"AllLabels\",\"id\":\"p1020\"}}}],\"right\":[{\"type\":\"object\",\"name\":\"Legend\",\"id\":\"p1087\",\"attributes\":{\"location\":\"center\",\"click_policy\":\"hide\",\"items\":[{\"type\":\"object\",\"name\":\"LegendItem\",\"id\":\"p1088\",\"attributes\":{\"label\":{\"type\":\"value\",\"value\":\"1975\"},\"renderers\":[{\"id\":\"p1039\"},{\"id\":\"p1057\"},{\"id\":\"p1048\"}]}},{\"type\":\"object\",\"name\":\"LegendItem\",\"id\":\"p1089\",\"attributes\":{\"label\":{\"type\":\"value\",\"value\":\"2012\"},\"renderers\":[{\"id\":\"p1066\"},{\"id\":\"p1084\"},{\"id\":\"p1075\"}]}}]}}],\"below\":[{\"type\":\"object\",\"name\":\"LinearAxis\",\"id\":\"p1012\",\"attributes\":{\"ticker\":{\"type\":\"object\",\"name\":\"BasicTicker\",\"id\":\"p1013\",\"attributes\":{\"mantissas\":[1,2,5]}},\"formatter\":{\"type\":\"object\",\"name\":\"BasicTickFormatter\",\"id\":\"p1014\"},\"axis_label\":\"beak depth (mm)\",\"major_label_policy\":{\"type\":\"object\",\"name\":\"AllLabels\",\"id\":\"p1015\"}}}],\"center\":[{\"type\":\"object\",\"name\":\"Grid\",\"id\":\"p1016\",\"attributes\":{\"axis\":{\"id\":\"p1012\"}}},{\"type\":\"object\",\"name\":\"Grid\",\"id\":\"p1021\",\"attributes\":{\"dimension\":1,\"axis\":{\"id\":\"p1017\"}}}],\"frame_width\":375,\"frame_height\":275}}]}};\n", " const render_items = [{\"docid\":\"6c137f60-05ae-4ef2-a7a9-00837278ad35\",\"roots\":{\"p1001\":\"a63d9a16-4307-4a7b-bf66-19822e6c9b90\"},\"root_ids\":[\"p1001\"]}];\n", " void root.Bokeh.embed.embed_items_notebook(docs_json, render_items);\n", " }\n", " if (root.Bokeh !== undefined) {\n", " embed_document(root);\n", " } else {\n", " let attempts = 0;\n", " const timer = setInterval(function(root) {\n", " if (root.Bokeh !== undefined) {\n", " clearInterval(timer);\n", " embed_document(root);\n", " } else {\n", " attempts++;\n", " if (attempts > 100) {\n", " clearInterval(timer);\n", " console.log(\"Bokeh: ERROR: Unable to run BokehJS code because BokehJS library is missing\");\n", " }\n", " }\n", " }, 10, root)\n", " }\n", "})(window);" ], "application/vnd.bokehjs_exec.v0+json": "" }, "metadata": { "application/vnd.bokehjs_exec.v0+json": { "id": "p1001" } }, "output_type": "display_data" } ], "source": [ "p = iqplot.ecdf(\n", " data=df,\n", " q='beak depth (mm)',\n", " cats='year',\n", ")\n", "\n", "bokeh.io.show(p)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Judging from the ECDFs, it seems as though beaks have gotten deeper over time. But now, we would like a *statistic* to compare. One statistic that comes to mind it the mean. So, let's compare those. First, we'll pull out the data sets as NumPy arrays for convenience (and speed later on when we start doing bootstrap replicates)." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(8.959999999999999, 9.188492063492063)" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# Pull our data sets as Numpy arrays\n", "bd_1975 = df.loc[df['year']==1975, 'beak depth (mm)'].values\n", "bd_2012 = df.loc[df['year']==2012, 'beak depth (mm)'].values\n", "\n", "# Compute the means\n", "np.mean(bd_1975), np.mean(bd_2012)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "So, indeed, the mean beak depth is bigger in 2012 than in 1975. But what if we did the measurements again under an identical set of conditions? Would we get similar results? To address this question, we would like to compute a *confidence interval* of the mean. We will compute the 95% confidence interval.\n", "\n", "What is a 95% confidence interval, in this case of a mean? It can be thought of as follows. If we were to repeat the experiment over and over and over again, 95% of the time, the observed mean would lie in the 95% confidence interval. So, if the confidence intervals of the means of measurements from 1975 and from 2012 overlapped, we might not be so sure that the beaks got deeper due to some underlying selective pressure, but that we just happened to *observe* deeper beaks as a result of natural variability.\n", "\n", "So, how do we compute a confidence interval? ....We use our computer!" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Bootstrap confidence intervals\n", "\n", "The notion of the bootstrap was first published by Brad Efron in 1979. The idea is simple, and we will take the fact that it works as a given; Efron proved it for us. \n", "\n", "Here's the idea: If we could somehow repeat the measurements of the beak depths on Daphne Major, we could do it many many times, and we could then just compute the 2.5th and 97.5th percentiles to get a 95% confidence interval. The problem is, we can't repeat the experiments over and over again. 1975 only happened once, and all birds on the island were measured. We cannot have 1975 happen again under exactly the same conditions. \n", "\n", "Instead, we will have our computer *simulate* doing the experiment over and over again. Hacker statistics! We have one set of measurements. We \"repeat\" the experiment by drawing measurements out of the ones we have again and again. Here's what we do to compute a bootstrap estimate of the mean of a set of _n_ data points.\n", "\n", "1. Draw _n_ data points out of the original data set *with replacement*. This set of data points is called a **bootstrap sample**.\n", "2. Compute the mean of the bootstrap sample. This is called a **bootstrap replicate** of the mean.\n", "3. Do this over and over again, storing the results.\n", "\n", "So, it is as if we did the experiment over and over again, obtaining a mean each time. Remember, our bootstrap sample has exactly the same number of \"measurements\" as the original data set. Let's try it with the `bd_1975` data (remember the mean beak depth was 8.96 mm). First we'll generate a bootstrap sample. Remember, the function `rng.choice()` allows us to sample out of an array with replacement, if we like." ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": [ "# Seed RNG for ease of commenting on values later;\n", "# generally do not seed the RNG when doing hacker stats\n", "rng = np.random.default_rng(3252)\n", "\n", "bs_sample = rng.choice(bd_1975, replace=True, size=len(bd_1975))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's take a quick look at this bootstrap sample by plotting its ECDF right next to that of the original data set." ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "
\n" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/javascript": [ "(function(root) {\n", " function embed_document(root) {\n", " const docs_json = {\"d61f84a3-ea86-4c5f-a2c1-9f1004d6afbc\":{\"version\":\"3.4.0\",\"title\":\"Bokeh Application\",\"roots\":[{\"type\":\"object\",\"name\":\"Figure\",\"id\":\"p1094\",\"attributes\":{\"x_range\":{\"type\":\"object\",\"name\":\"DataRange1d\",\"id\":\"p1095\"},\"y_range\":{\"type\":\"object\",\"name\":\"DataRange1d\",\"id\":\"p1096\"},\"x_scale\":{\"type\":\"object\",\"name\":\"LinearScale\",\"id\":\"p1103\"},\"y_scale\":{\"type\":\"object\",\"name\":\"LinearScale\",\"id\":\"p1104\"},\"title\":{\"type\":\"object\",\"name\":\"Title\",\"id\":\"p1101\"},\"renderers\":[{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1132\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1126\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1127\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1128\"},\"data\":{\"type\":\"map\",\"entries\":[[\"x\",{\"type\":\"ndarray\",\"array\":{\"type\":\"bytes\",\"data\":\"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\"},\"shape\":[174],\"dtype\":\"float64\",\"order\":\"little\"}],[\"y\",{\"type\":\"ndarray\",\"array\":{\"type\":\"bytes\",\"data\":\"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\"},\"shape\":[174],\"dtype\":\"float64\",\"order\":\"little\"}]]}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1133\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1134\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1129\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#1f77b3\",\"line_width\":2}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1130\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#1f77b3\",\"line_alpha\":0.1,\"line_width\":2}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1131\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#1f77b3\",\"line_alpha\":0.2,\"line_width\":2}}}},{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1141\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1135\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1136\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1137\"},\"data\":{\"type\":\"map\"}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1142\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1143\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1138\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":7.9},\"y\":{\"type\":\"value\",\"value\":0},\"angle\":{\"type\":\"value\",\"value\":3.141592653589793},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1139\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":7.9},\"y\":{\"type\":\"value\",\"value\":0},\"angle\":{\"type\":\"value\",\"value\":3.141592653589793},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.1},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1140\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":7.9},\"y\":{\"type\":\"value\",\"value\":0},\"angle\":{\"type\":\"value\",\"value\":3.141592653589793},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.2},\"line_width\":{\"type\":\"value\",\"value\":2}}}}},{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1150\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1144\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1145\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1146\"},\"data\":{\"type\":\"map\"}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1151\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1152\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1147\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":10.4},\"y\":{\"type\":\"value\",\"value\":1},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1148\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":10.4},\"y\":{\"type\":\"value\",\"value\":1},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.1},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1149\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":10.4},\"y\":{\"type\":\"value\",\"value\":1},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.2},\"line_width\":{\"type\":\"value\",\"value\":2}}}}},{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1159\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1153\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1154\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1155\"},\"data\":{\"type\":\"map\",\"entries\":[[\"x\",{\"type\":\"ndarray\",\"array\":{\"type\":\"bytes\",\"data\":\"mpmZmZmZH0CamZmZmZkfQJqZmZmZmR9AmpmZmZmZH0AAAAAAAAAgQAAAAAAAACBAAAAAAAAAIEAAAAAAAAAgQAAAAAAAACBAAAAAAAAAIEAAAAAAAAAgQAAAAAAAACBAAAAAAAAAIEAAAAAAAAAgQAAAAAAAACBAAAAAAAAAIEBmZmZmZmYgQGZmZmZmZiBAZmZmZmZmIEBmZmZmZmYgQJqZmZmZmSBAmpmZmZmZIECamZmZmZkgQJqZmZmZmSBAmpmZmZmZIECamZmZmZkgQJqZmZmZmSBAmpmZmZmZIECamZmZmZkgQJqZmZmZmSBAzczMzMzMIEDNzMzMzMwgQM3MzMzMzCBAzczMzMzMIEDNzMzMzMwgQM3MzMzMzCBAzczMzMzMIEDNzMzMzMwgQM3MzMzMzCBAzczMzMzMIEDNzMzMzMwgQM3MzMzMzCBAAAAAAAAAIUAAAAAAAAAhQAAAAAAAACFAAAAAAAAAIUAAAAAAAAAhQAAAAAAAACFAAAAAAAAAIUAAAAAAAAAhQAAAAAAAACFAAAAAAAAAIUAAAAAAAAAhQAAAAAAAACFAAAAAAAAAIUAAAAAAAAAhQAAAAAAAACFAAAAAAAAAIUAzMzMzMzMhQDMzMzMzMyFAMzMzMzMzIUAzMzMzMzMhQDMzMzMzMyFAMzMzMzMzIUAzMzMzMzMhQDMzMzMzMyFAMzMzMzMzIUAzMzMzMzMhQDMzMzMzMyFAMzMzMzMzIUBmZmZmZmYhQGZmZmZmZiFAZmZmZmZmIUBmZmZmZmYhQGZmZmZmZiFAZmZmZmZmIUCamZmZmZkhQJqZmZmZmSFAmpmZmZmZIUCamZmZmZkhQM3MzMzMzCFAzczMzMzMIUDNzMzMzMwhQM3MzMzMzCFAzczMzMzMIUDNzMzMzMwhQAAAAAAAACJAAAAAAAAAIkAAAAAAAAAiQAAAAAAAACJAAAAAAAAAIkAAAAAAAAAiQAAAAAAAACJAAAAAAAAAIkAAAAAAAAAiQAAAAAAAACJAAAAAAAAAIkAAAAAAAAAiQBSuR+F6FCJAFK5H4XoUIkAUrkfhehQiQBSuR+F6FCJAMzMzMzMzIkAzMzMzMzMiQDMzMzMzMyJAMzMzMzMzIkAzMzMzMzMiQDMzMzMzMyJAMzMzMzMzIkAzMzMzMzMiQDMzMzMzMyJAMzMzMzMzIkAzMzMzMzMiQDMzMzMzMyJAMzMzMzMzIkAzMzMzMzMiQDMzMzMzMyJAMzMzMzMzIkAzMzMzMzMiQDMzMzMzMyJAMzMzMzMzIkAzMzMzMzMiQDMzMzMzMyJAMzMzMzMzIkAzMzMzMzMiQDMzMzMzMyJAZmZmZmZmIkBmZmZmZmYiQGZmZmZmZiJAZmZmZmZmIkBmZmZmZmYiQGZmZmZmZiJAZmZmZmZmIkBmZmZmZmYiQGZmZmZmZiJAZmZmZmZmIkBmZmZmZmYiQGZmZmZmZiJAZmZmZmZmIkBmZmZmZmYiQGZmZmZmZiJAZmZmZmZmIkBmZmZmZmYiQGZmZmZmZiJAmpmZmZmZIkCamZmZmZkiQJqZmZmZmSJAmpmZmZmZIkBmZmZmZuYiQGZmZmZm5iJAZmZmZmbmIkBmZmZmZuYiQGZmZmZm5iJAZmZmZmbmIkBmZmZmZuYiQGZmZmZm5iJAZmZmZmbmIkBmZmZmZuYiQGZmZmZm5iJAZmZmZmbmIkAzMzMzMzMjQDMzMzMzMyNAMzMzMzMzI0AzMzMzMzMjQDMzMzMzMyNAMzMzMzMzI0DNzMzMzEwjQM3MzMzMTCNAmpmZmZmZI0CamZmZmZkjQJqZmZmZmSNAmpmZmZmZI0DNzMzMzMwkQM3MzMzMzCRA\"},\"shape\":[174],\"dtype\":\"float64\",\"order\":\"little\"}],[\"y\",{\"type\":\"ndarray\",\"array\":{\"type\":\"bytes\",\"data\":\"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\"},\"shape\":[174],\"dtype\":\"float64\",\"order\":\"little\"}]]}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1160\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1161\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1156\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"gray\",\"line_width\":2}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1157\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"gray\",\"line_alpha\":0.1,\"line_width\":2}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1158\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"gray\",\"line_alpha\":0.2,\"line_width\":2}}}},{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1168\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1162\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1163\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1164\"},\"data\":{\"type\":\"map\"}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1169\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1170\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1165\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":7.9},\"y\":{\"type\":\"value\",\"value\":0},\"angle\":{\"type\":\"value\",\"value\":3.141592653589793},\"line_color\":{\"type\":\"value\",\"value\":\"gray\"},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1166\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":7.9},\"y\":{\"type\":\"value\",\"value\":0},\"angle\":{\"type\":\"value\",\"value\":3.141592653589793},\"line_color\":{\"type\":\"value\",\"value\":\"gray\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.1},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1167\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":7.9},\"y\":{\"type\":\"value\",\"value\":0},\"angle\":{\"type\":\"value\",\"value\":3.141592653589793},\"line_color\":{\"type\":\"value\",\"value\":\"gray\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.2},\"line_width\":{\"type\":\"value\",\"value\":2}}}}},{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1177\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1171\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1172\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1173\"},\"data\":{\"type\":\"map\"}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1178\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1179\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1174\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":10.4},\"y\":{\"type\":\"value\",\"value\":1},\"line_color\":{\"type\":\"value\",\"value\":\"gray\"},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1175\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":10.4},\"y\":{\"type\":\"value\",\"value\":1},\"line_color\":{\"type\":\"value\",\"value\":\"gray\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.1},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1176\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":10.4},\"y\":{\"type\":\"value\",\"value\":1},\"line_color\":{\"type\":\"value\",\"value\":\"gray\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.2},\"line_width\":{\"type\":\"value\",\"value\":2}}}}}],\"toolbar\":{\"type\":\"object\",\"name\":\"Toolbar\",\"id\":\"p1102\",\"attributes\":{\"tools\":[{\"type\":\"object\",\"name\":\"PanTool\",\"id\":\"p1115\"},{\"type\":\"object\",\"name\":\"WheelZoomTool\",\"id\":\"p1116\",\"attributes\":{\"renderers\":\"auto\"}},{\"type\":\"object\",\"name\":\"BoxZoomTool\",\"id\":\"p1117\",\"attributes\":{\"overlay\":{\"type\":\"object\",\"name\":\"BoxAnnotation\",\"id\":\"p1118\",\"attributes\":{\"syncable\":false,\"level\":\"overlay\",\"visible\":false,\"left\":{\"type\":\"number\",\"value\":\"nan\"},\"right\":{\"type\":\"number\",\"value\":\"nan\"},\"top\":{\"type\":\"number\",\"value\":\"nan\"},\"bottom\":{\"type\":\"number\",\"value\":\"nan\"},\"left_units\":\"canvas\",\"right_units\":\"canvas\",\"top_units\":\"canvas\",\"bottom_units\":\"canvas\",\"line_color\":\"black\",\"line_alpha\":1.0,\"line_width\":2,\"line_dash\":[4,4],\"fill_color\":\"lightgrey\",\"fill_alpha\":0.5}}}},{\"type\":\"object\",\"name\":\"SaveTool\",\"id\":\"p1123\"},{\"type\":\"object\",\"name\":\"ResetTool\",\"id\":\"p1124\"},{\"type\":\"object\",\"name\":\"HelpTool\",\"id\":\"p1125\"}]}},\"toolbar_location\":\"above\",\"left\":[{\"type\":\"object\",\"name\":\"LinearAxis\",\"id\":\"p1110\",\"attributes\":{\"ticker\":{\"type\":\"object\",\"name\":\"BasicTicker\",\"id\":\"p1111\",\"attributes\":{\"mantissas\":[1,2,5]}},\"formatter\":{\"type\":\"object\",\"name\":\"BasicTickFormatter\",\"id\":\"p1112\"},\"axis_label\":\"ECDF\",\"major_label_policy\":{\"type\":\"object\",\"name\":\"AllLabels\",\"id\":\"p1113\"}}}],\"below\":[{\"type\":\"object\",\"name\":\"LinearAxis\",\"id\":\"p1105\",\"attributes\":{\"ticker\":{\"type\":\"object\",\"name\":\"BasicTicker\",\"id\":\"p1106\",\"attributes\":{\"mantissas\":[1,2,5]}},\"formatter\":{\"type\":\"object\",\"name\":\"BasicTickFormatter\",\"id\":\"p1107\"},\"axis_label\":\"beak depth (mm)\",\"major_label_policy\":{\"type\":\"object\",\"name\":\"AllLabels\",\"id\":\"p1108\"}}}],\"center\":[{\"type\":\"object\",\"name\":\"Grid\",\"id\":\"p1109\",\"attributes\":{\"axis\":{\"id\":\"p1105\"}}},{\"type\":\"object\",\"name\":\"Grid\",\"id\":\"p1114\",\"attributes\":{\"dimension\":1,\"axis\":{\"id\":\"p1110\"}}}],\"frame_width\":375,\"frame_height\":275}}]}};\n", " const render_items = [{\"docid\":\"d61f84a3-ea86-4c5f-a2c1-9f1004d6afbc\",\"roots\":{\"p1094\":\"bd2baba7-8b33-4018-b9a5-c0aa1eadf87e\"},\"root_ids\":[\"p1094\"]}];\n", " void root.Bokeh.embed.embed_items_notebook(docs_json, render_items);\n", " }\n", " if (root.Bokeh !== undefined) {\n", " embed_document(root);\n", " } else {\n", " let attempts = 0;\n", " const timer = setInterval(function(root) {\n", " if (root.Bokeh !== undefined) {\n", " clearInterval(timer);\n", " embed_document(root);\n", " } else {\n", " attempts++;\n", " if (attempts > 100) {\n", " clearInterval(timer);\n", " console.log(\"Bokeh: ERROR: Unable to run BokehJS code because BokehJS library is missing\");\n", " }\n", " }\n", " }, 10, root)\n", " }\n", "})(window);" ], "application/vnd.bokehjs_exec.v0+json": "" }, "metadata": { "application/vnd.bokehjs_exec.v0+json": { "id": "p1094" } }, "output_type": "display_data" } ], "source": [ "# Original data set\n", "p = iqplot.ecdf(\n", " data=df.loc[df['year']==1975, :],\n", " q='beak depth (mm)',\n", ")\n", "\n", "# Bootstrap data set\n", "p = iqplot.ecdf(\n", " data=bs_sample,\n", " q='bootstrap',\n", " cats=None,\n", " p=p,\n", " line_kwargs=dict(\n", " line_color='gray'\n", " ),\n", ")\n", "\n", "bokeh.io.show(p)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "It is qualitatively similar, but obviously not exactly the same data set.\n", "\n", "Now, let's compute our bootstrap replicate. It's as simple as computing the mean of the bootstrap sample." ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "8.849770114942531" ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "bs_replicate = np.mean(bs_sample)\n", "bs_replicate" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "So, the mean of the bootstrap replicate is 8.84 mm, which is less than the mean of 8.96 from the original data set.\n", "\n", "Now, we can write a `for` loop to get lots and lots of bootstrap replicates. Note the since you are doing the replicates many many times, speed matters. For this reason, be sure you convert the data you are bootstrapping into a Numpy array. The calculations with them are **much** faster than with Pandas `Series`." ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [], "source": [ "# Number of replicatess\n", "n_reps = 2000\n", "\n", "# Initialize bootstrap replicas array\n", "bs_reps_1975 = np.empty(n_reps)\n", "\n", "# Compute replicates\n", "for i in range(n_reps):\n", " bs_sample = rng.choice(bd_1975, size=len(bd_1975))\n", " bs_reps_1975[i] = np.mean(bs_sample)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Now that we have our replicas, 2,000 of them, we can plot an ECDF to see what we might expect of the mean if we were to do the experiment again." ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "
\n" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/javascript": [ "(function(root) {\n", " function embed_document(root) {\n", " const docs_json = {\"19af452f-d830-4c51-808d-057e3c3b33e0\":{\"version\":\"3.4.0\",\"title\":\"Bokeh Application\",\"roots\":[{\"type\":\"object\",\"name\":\"Figure\",\"id\":\"p1184\",\"attributes\":{\"x_range\":{\"type\":\"object\",\"name\":\"DataRange1d\",\"id\":\"p1185\"},\"y_range\":{\"type\":\"object\",\"name\":\"DataRange1d\",\"id\":\"p1186\"},\"x_scale\":{\"type\":\"object\",\"name\":\"LinearScale\",\"id\":\"p1193\"},\"y_scale\":{\"type\":\"object\",\"name\":\"LinearScale\",\"id\":\"p1194\"},\"title\":{\"type\":\"object\",\"name\":\"Title\",\"id\":\"p1191\"},\"renderers\":[{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1222\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1216\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1217\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1218\"},\"data\":{\"type\":\"map\",\"entries\":[[\"x\",{\"type\":\"ndarray\",\"array\":{\"type\":\"bytes\",\"data\":\"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\"},\"shape\":[4000],\"dtype\":\"float64\",\"order\":\"little\"}],[\"y\",{\"type\":\"ndarray\",\"array\":{\"type\":\"bytes\",\"data\":\"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\"},\"shape\":[4000],\"dtype\":\"float64\",\"order\":\"little\"}]]}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1223\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1224\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1219\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#1f77b3\",\"line_width\":2}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1220\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#1f77b3\",\"line_alpha\":0.1,\"line_width\":2}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1221\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#1f77b3\",\"line_alpha\":0.2,\"line_width\":2}}}},{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1231\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1225\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1226\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1227\"},\"data\":{\"type\":\"map\"}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1232\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1233\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1228\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":8.76448275862069},\"y\":{\"type\":\"value\",\"value\":0},\"angle\":{\"type\":\"value\",\"value\":3.141592653589793},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1229\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":8.76448275862069},\"y\":{\"type\":\"value\",\"value\":0},\"angle\":{\"type\":\"value\",\"value\":3.141592653589793},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.1},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1230\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":8.76448275862069},\"y\":{\"type\":\"value\",\"value\":0},\"angle\":{\"type\":\"value\",\"value\":3.141592653589793},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.2},\"line_width\":{\"type\":\"value\",\"value\":2}}}}},{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1240\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1234\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1235\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1236\"},\"data\":{\"type\":\"map\"}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1241\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1242\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1237\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":9.172068965517242},\"y\":{\"type\":\"value\",\"value\":1},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1238\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":9.172068965517242},\"y\":{\"type\":\"value\",\"value\":1},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.1},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1239\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":9.172068965517242},\"y\":{\"type\":\"value\",\"value\":1},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.2},\"line_width\":{\"type\":\"value\",\"value\":2}}}}}],\"toolbar\":{\"type\":\"object\",\"name\":\"Toolbar\",\"id\":\"p1192\",\"attributes\":{\"tools\":[{\"type\":\"object\",\"name\":\"PanTool\",\"id\":\"p1205\"},{\"type\":\"object\",\"name\":\"WheelZoomTool\",\"id\":\"p1206\",\"attributes\":{\"renderers\":\"auto\"}},{\"type\":\"object\",\"name\":\"BoxZoomTool\",\"id\":\"p1207\",\"attributes\":{\"overlay\":{\"type\":\"object\",\"name\":\"BoxAnnotation\",\"id\":\"p1208\",\"attributes\":{\"syncable\":false,\"level\":\"overlay\",\"visible\":false,\"left\":{\"type\":\"number\",\"value\":\"nan\"},\"right\":{\"type\":\"number\",\"value\":\"nan\"},\"top\":{\"type\":\"number\",\"value\":\"nan\"},\"bottom\":{\"type\":\"number\",\"value\":\"nan\"},\"left_units\":\"canvas\",\"right_units\":\"canvas\",\"top_units\":\"canvas\",\"bottom_units\":\"canvas\",\"line_color\":\"black\",\"line_alpha\":1.0,\"line_width\":2,\"line_dash\":[4,4],\"fill_color\":\"lightgrey\",\"fill_alpha\":0.5}}}},{\"type\":\"object\",\"name\":\"SaveTool\",\"id\":\"p1213\"},{\"type\":\"object\",\"name\":\"ResetTool\",\"id\":\"p1214\"},{\"type\":\"object\",\"name\":\"HelpTool\",\"id\":\"p1215\"}]}},\"toolbar_location\":\"above\",\"left\":[{\"type\":\"object\",\"name\":\"LinearAxis\",\"id\":\"p1200\",\"attributes\":{\"ticker\":{\"type\":\"object\",\"name\":\"BasicTicker\",\"id\":\"p1201\",\"attributes\":{\"mantissas\":[1,2,5]}},\"formatter\":{\"type\":\"object\",\"name\":\"BasicTickFormatter\",\"id\":\"p1202\"},\"axis_label\":\"ECDF\",\"major_label_policy\":{\"type\":\"object\",\"name\":\"AllLabels\",\"id\":\"p1203\"}}}],\"below\":[{\"type\":\"object\",\"name\":\"LinearAxis\",\"id\":\"p1195\",\"attributes\":{\"ticker\":{\"type\":\"object\",\"name\":\"BasicTicker\",\"id\":\"p1196\",\"attributes\":{\"mantissas\":[1,2,5]}},\"formatter\":{\"type\":\"object\",\"name\":\"BasicTickFormatter\",\"id\":\"p1197\"},\"axis_label\":\"mean beak depth (mm)\",\"major_label_policy\":{\"type\":\"object\",\"name\":\"AllLabels\",\"id\":\"p1198\"}}}],\"center\":[{\"type\":\"object\",\"name\":\"Grid\",\"id\":\"p1199\",\"attributes\":{\"axis\":{\"id\":\"p1195\"}}},{\"type\":\"object\",\"name\":\"Grid\",\"id\":\"p1204\",\"attributes\":{\"dimension\":1,\"axis\":{\"id\":\"p1200\"}}}],\"frame_width\":375,\"frame_height\":275}}]}};\n", " const render_items = [{\"docid\":\"19af452f-d830-4c51-808d-057e3c3b33e0\",\"roots\":{\"p1184\":\"b1d054bd-8e03-4451-8bf3-78eb6cba9699\"},\"root_ids\":[\"p1184\"]}];\n", " void root.Bokeh.embed.embed_items_notebook(docs_json, render_items);\n", " }\n", " if (root.Bokeh !== undefined) {\n", " embed_document(root);\n", " } else {\n", " let attempts = 0;\n", " const timer = setInterval(function(root) {\n", " if (root.Bokeh !== undefined) {\n", " clearInterval(timer);\n", " embed_document(root);\n", " } else {\n", " attempts++;\n", " if (attempts > 100) {\n", " clearInterval(timer);\n", " console.log(\"Bokeh: ERROR: Unable to run BokehJS code because BokehJS library is missing\");\n", " }\n", " }\n", " }, 10, root)\n", " }\n", "})(window);" ], "application/vnd.bokehjs_exec.v0+json": "" }, "metadata": { "application/vnd.bokehjs_exec.v0+json": { "id": "p1184" } }, "output_type": "display_data" } ], "source": [ "# Original data set\n", "p = iqplot.ecdf(\n", " bs_reps_1975,\n", " x_axis_label='mean beak depth (mm)',\n", ")\n", "\n", "bokeh.io.show(p)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "It looks Normally distributed, and in fact it must be approximately Normally distributed by the Central Limit Theorem (which we will not discuss here, but we didn't really need to derive; hacker statistics brought us here!). The most probable mean (located at the inflection point on the CDF) we would get is 8.96 mm, which is what was measured, but upon repeating the experiment, we could get a mean as low as about 8.75 mm or as high as about 9.2 mm.\n", "\n", "Let's compute a 95% confidence interval. We use `np.percentile()` with the bootstrap replicates to compute the 2.5th and 97.5th percentiles to give a 95% confidence interval." ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([8.8466523 , 9.07979023])" ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "conf_int_1975 = np.percentile(bs_reps_1975, [2.5, 97.5])\n", "conf_int_1975" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### A function for replicates\n", "\n", "The construction we had for making our bootstrap replicates was a bit clunky:\n", "\n", "```python\n", "# Initialize bootstrap replicas array\n", "bs_reps_1975 = np.empty(n_reps)\n", "\n", "# Compute replicates\n", "for i in range(n_reps):\n", " bs_sample = rng.choice(bd_1975, size=len(bd_1975))\n", " bs_reps_1975[i] = np.mean(bs_sample)\n", "```\n", "\n", "We had to set up an empty array, and then loop through each index, draw a bootstrap sample, compute its mean to get the replicate, and then place it in the array. We could, instead, write a function to compute a bootstrap replicate from data." ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [], "source": [ "def draw_bs_rep(data, func, rng):\n", " \"\"\"Compute a bootstrap replicate from data.\"\"\"\n", " bs_sample = rng.choice(data, size=len(data))\n", " return func(bs_sample)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Note that this function is generic in that it can compute the replicate using any summary statistic function, such as `np.median()`, `np.std()`, or anything else. With this function in hand, our code starts to look a little cleaner.\n", "\n", "```python\n", "# Initialize bootstrap replicas array\n", "bs_reps_1975 = np.empty(n_reps)\n", "\n", "# Compute replicates\n", "for i in range(n_reps):\n", " bs_reps_1975 = draw_bs_rep(bd_1975, np.mean, rng)\n", "```" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We can use this function in a list comprehension to quickly get an array of replicates." ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [], "source": [ "bs_reps_1975 = np.array(\n", " [draw_bs_rep(bd_1975, np.mean, rng) for _ in range(n_reps)]\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This is much more concise and perhaps cleaner syntax. Now let's use this construction to make bootstrap replicates for the 2012 samples." ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [], "source": [ "# Compute replicates\n", "bs_reps_2012 = np.array(\n", " [draw_bs_rep(bd_2012, np.mean, rng) for _ in range(n_reps)]\n", ")\n", "\n", "# Compute the confidence interval\n", "conf_int_2012 = np.percentile(bs_reps_2012, [2.5, 97.5])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We can print the two confidence intervals next to each other for comparison." ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[8.8466523 9.07979023]\n", "[9.07458333 9.30000992]\n" ] } ], "source": [ "print(conf_int_1975)\n", "print(conf_int_2012)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "So, the 95% confidence intervals for the 2012 and 1975 juuust overlap. This implies that the inherent variation in beak depths is likely not responsible for the observed difference. There was likely some selective pressure toward deeper beaks." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Plotting and reporting confidence intervals\n", "\n", "In the above code cell, we have shown numerical values for the edges of the 95% confidence interval for beak depth in 1975 and 2012. How would you display these results in a publication?\n", "\n", "We will start by discussing how you would express confidence intervals in text. You may have seen confidence intervals expressed like this: 10.3 ± 2.1. I never write my confidence intervals this way because *confidence intervals are in general not symmetric*. Rather, would report the confidence intervals for the above calculations like this:\n", "\n", "- 1975 beak depth: $8.96^{+0.11}_{-0.12}$ mm.\n", "- 2012 beak depth: $9.19^{+0.11}_{-0.11}$ mm.\n", "\n", "Better yet, we can show the confidence intervals graphically. I will make a plot and then discuss my design choices and some of the details about how I made the plot." ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [ { "data": { "text/html": [ "\n", "
\n" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/javascript": [ "(function(root) {\n", " function embed_document(root) {\n", " const docs_json = {\"5d644e9d-cfab-4ad2-a00c-7f017a55ebb1\":{\"version\":\"3.4.0\",\"title\":\"Bokeh Application\",\"roots\":[{\"type\":\"object\",\"name\":\"Figure\",\"id\":\"p1245\",\"attributes\":{\"x_range\":{\"type\":\"object\",\"name\":\"DataRange1d\",\"id\":\"p1247\"},\"y_range\":{\"type\":\"object\",\"name\":\"FactorRange\",\"id\":\"p1254\",\"attributes\":{\"factors\":[\"2012\",\"1975\"]}},\"x_scale\":{\"type\":\"object\",\"name\":\"LinearScale\",\"id\":\"p1255\"},\"y_scale\":{\"type\":\"object\",\"name\":\"CategoricalScale\",\"id\":\"p1256\"},\"title\":{\"type\":\"object\",\"name\":\"Title\",\"id\":\"p1252\"},\"renderers\":[{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1284\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1278\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1279\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1280\"},\"data\":{\"type\":\"map\",\"entries\":[[\"x\",[9.188492063492063,8.959999999999999]],[\"y\",[\"2012\",\"1975\"]]]}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1285\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1286\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Scatter\",\"id\":\"p1281\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"size\":{\"type\":\"value\",\"value\":5},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b4\"},\"fill_color\":{\"type\":\"value\",\"value\":\"#1f77b4\"}}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Scatter\",\"id\":\"p1282\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"size\":{\"type\":\"value\",\"value\":5},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b4\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.1},\"fill_color\":{\"type\":\"value\",\"value\":\"#1f77b4\"},\"fill_alpha\":{\"type\":\"value\",\"value\":0.1},\"hatch_alpha\":{\"type\":\"value\",\"value\":0.1}}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Scatter\",\"id\":\"p1283\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"size\":{\"type\":\"value\",\"value\":5},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b4\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.2},\"fill_color\":{\"type\":\"value\",\"value\":\"#1f77b4\"},\"fill_alpha\":{\"type\":\"value\",\"value\":0.2},\"hatch_alpha\":{\"type\":\"value\",\"value\":0.2}}}}},{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1293\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1287\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1288\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1289\"},\"data\":{\"type\":\"map\",\"entries\":[[\"x\",{\"type\":\"ndarray\",\"array\":{\"type\":\"bytes\",\"data\":\"+0r9aHyxIUAVy9dD2igiQA==\"},\"shape\":[2],\"dtype\":\"float64\",\"order\":\"little\"}],[\"y\",[\"1975\",\"1975\"]]]}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1294\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1295\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1290\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#1f77b4\",\"line_width\":3}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1291\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#1f77b4\",\"line_alpha\":0.1,\"line_width\":3}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1292\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#1f77b4\",\"line_alpha\":0.2,\"line_width\":3}}}},{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1302\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1296\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1297\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1298\"},\"data\":{\"type\":\"map\",\"entries\":[[\"x\",{\"type\":\"ndarray\",\"array\":{\"type\":\"bytes\",\"data\":\"l/xiyS8mIkCANHvmmpkiQA==\"},\"shape\":[2],\"dtype\":\"float64\",\"order\":\"little\"}],[\"y\",[\"2012\",\"2012\"]]]}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1303\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1304\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1299\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#1f77b4\",\"line_width\":3}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1300\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#1f77b4\",\"line_alpha\":0.1,\"line_width\":3}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1301\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#1f77b4\",\"line_alpha\":0.2,\"line_width\":3}}}}],\"toolbar\":{\"type\":\"object\",\"name\":\"Toolbar\",\"id\":\"p1253\",\"attributes\":{\"tools\":[{\"type\":\"object\",\"name\":\"PanTool\",\"id\":\"p1267\"},{\"type\":\"object\",\"name\":\"WheelZoomTool\",\"id\":\"p1268\",\"attributes\":{\"renderers\":\"auto\"}},{\"type\":\"object\",\"name\":\"BoxZoomTool\",\"id\":\"p1269\",\"attributes\":{\"overlay\":{\"type\":\"object\",\"name\":\"BoxAnnotation\",\"id\":\"p1270\",\"attributes\":{\"syncable\":false,\"level\":\"overlay\",\"visible\":false,\"left\":{\"type\":\"number\",\"value\":\"nan\"},\"right\":{\"type\":\"number\",\"value\":\"nan\"},\"top\":{\"type\":\"number\",\"value\":\"nan\"},\"bottom\":{\"type\":\"number\",\"value\":\"nan\"},\"left_units\":\"canvas\",\"right_units\":\"canvas\",\"top_units\":\"canvas\",\"bottom_units\":\"canvas\",\"line_color\":\"black\",\"line_alpha\":1.0,\"line_width\":2,\"line_dash\":[4,4],\"fill_color\":\"lightgrey\",\"fill_alpha\":0.5}}}},{\"type\":\"object\",\"name\":\"SaveTool\",\"id\":\"p1275\"},{\"type\":\"object\",\"name\":\"ResetTool\",\"id\":\"p1276\"},{\"type\":\"object\",\"name\":\"HelpTool\",\"id\":\"p1277\"}]}},\"left\":[{\"type\":\"object\",\"name\":\"CategoricalAxis\",\"id\":\"p1262\",\"attributes\":{\"ticker\":{\"type\":\"object\",\"name\":\"CategoricalTicker\",\"id\":\"p1263\"},\"formatter\":{\"type\":\"object\",\"name\":\"CategoricalTickFormatter\",\"id\":\"p1264\"},\"major_label_policy\":{\"type\":\"object\",\"name\":\"AllLabels\",\"id\":\"p1265\"}}}],\"below\":[{\"type\":\"object\",\"name\":\"LinearAxis\",\"id\":\"p1257\",\"attributes\":{\"ticker\":{\"type\":\"object\",\"name\":\"BasicTicker\",\"id\":\"p1258\",\"attributes\":{\"mantissas\":[1,2,5]}},\"formatter\":{\"type\":\"object\",\"name\":\"BasicTickFormatter\",\"id\":\"p1259\"},\"axis_label\":\"beak depth (mm)\",\"major_label_policy\":{\"type\":\"object\",\"name\":\"AllLabels\",\"id\":\"p1260\"}}}],\"center\":[{\"type\":\"object\",\"name\":\"Grid\",\"id\":\"p1261\",\"attributes\":{\"axis\":{\"id\":\"p1257\"}}},{\"type\":\"object\",\"name\":\"Grid\",\"id\":\"p1266\",\"attributes\":{\"dimension\":1,\"axis\":{\"id\":\"p1262\"}}}],\"frame_width\":250,\"frame_height\":100}}]}};\n", " const render_items = [{\"docid\":\"5d644e9d-cfab-4ad2-a00c-7f017a55ebb1\",\"roots\":{\"p1245\":\"c9acd7b7-018a-4028-ad1b-618127c3b20a\"},\"root_ids\":[\"p1245\"]}];\n", " void root.Bokeh.embed.embed_items_notebook(docs_json, render_items);\n", " }\n", " if (root.Bokeh !== undefined) {\n", " embed_document(root);\n", " } else {\n", " let attempts = 0;\n", " const timer = setInterval(function(root) {\n", " if (root.Bokeh !== undefined) {\n", " clearInterval(timer);\n", " embed_document(root);\n", " } else {\n", " attempts++;\n", " if (attempts > 100) {\n", " clearInterval(timer);\n", " console.log(\"Bokeh: ERROR: Unable to run BokehJS code because BokehJS library is missing\");\n", " }\n", " }\n", " }, 10, root)\n", " }\n", "})(window);" ], "application/vnd.bokehjs_exec.v0+json": "" }, "metadata": { "application/vnd.bokehjs_exec.v0+json": { "id": "p1245" } }, "output_type": "display_data" } ], "source": [ "years = ['2012', '1975']\n", "p = bokeh.plotting.figure(\n", " frame_height=100,\n", " frame_width=250,\n", " x_axis_label='beak depth (mm)',\n", " y_range=years,\n", ")\n", "\n", "p.scatter([bd_2012.mean(), bd_1975.mean()], years, marker=\"circle\", size=5)\n", "p.line(conf_int_1975, ['1975']*2, line_width=3)\n", "p.line(conf_int_2012, ['2012']*2, line_width=3)\n", "\n", "bokeh.io.show(p)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "I prefer to make my plots of confidence intervals with error bars oriented horizontally. Most plots I see have them oriented vertically, but I find this counterintuitive. If we are thinking of the confidence interval as a summary of a probability density function or of a cumulative distribution function, the x-axis should have the measured value. Furthermore, the other axis is usually categorical, so it often contains text that is to be read. In this case, they are just years, but in many applications they are things like genotypes, treatments, etc. These are easier to read if they are on the vertical axis because they can easily be read horizontally without rotating text.\n", "\n", "To make this plot, I made a categorical y-axis with Bokeh. To do this, when I set up the figure, I specified the `y_range` kwargs to be a list of strings. This tells Bokeh that the axis is categorical. Then, when I populated the glyphs, I simply enter the values of the strings as the y-axis values." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Equivalence of standard deviation bootstrap samples of the mean and standard error of the mean\n", "\n", "The **standard error of the mean**, or SEM, is a measure of uncertainty of the estimate of the mean. In other words, if we did the set of measurements again, we would get a different mean. The variability in these measured means is described by the SEM. Specifically, it is the standard deviation of the Normal distribution describing the mean of repeated measurements. So, from bootstrap replicates, we can directly apply this formula." ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.06003142350273925" ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" } ], "source": [ "bs_sem = np.std(bs_reps_1975)\n", "bs_sem" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "It can be shown analytically that the SEM can be computed directly from the measurements as the standard deviation of the measurements divided by the square root of the number of measurements." ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "0.06074539219629801" ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" } ], "source": [ "sem = np.std(bd_1975, ddof=1) / np.sqrt(len(bd_1975))\n", "sem" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Hey, we got the same result! You may think I'm a jerk for making you get a simple answer the hard way by bootstrapping. But remember that bootstrap replicates are easy to generate in general for *any* statistic, and the confidence intervals on those statistics might not be as simple as for the mean, as we will demonstrate now." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Bootstrap confidence interval of the standard deviation\n", "\n", "We are not limited to computing bootstrap confidence intervals of the mean. We could compute bootstrap confidence intervals of any statistic, like the median, standard deviation, the standard deviation divided by the mean (coefficient of variation), whatever we like. Computing the confidence interval for the standard deviation is the same procedure as we have done; we just put `np.std` in for `np.mean`." ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[0.47835178 0.63827714]\n" ] }, { "data": { "text/html": [ "\n", "
\n" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/javascript": [ "(function(root) {\n", " function embed_document(root) {\n", " const docs_json = {\"0b2e7d1b-00a1-4956-8235-e68a9cd87b0a\":{\"version\":\"3.4.0\",\"title\":\"Bokeh Application\",\"roots\":[{\"type\":\"object\",\"name\":\"Figure\",\"id\":\"p1307\",\"attributes\":{\"x_range\":{\"type\":\"object\",\"name\":\"DataRange1d\",\"id\":\"p1308\"},\"y_range\":{\"type\":\"object\",\"name\":\"DataRange1d\",\"id\":\"p1309\"},\"x_scale\":{\"type\":\"object\",\"name\":\"LinearScale\",\"id\":\"p1316\"},\"y_scale\":{\"type\":\"object\",\"name\":\"LinearScale\",\"id\":\"p1317\"},\"title\":{\"type\":\"object\",\"name\":\"Title\",\"id\":\"p1314\"},\"renderers\":[{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1345\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1339\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1340\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1341\"},\"data\":{\"type\":\"map\",\"entries\":[[\"x\",{\"type\":\"ndarray\",\"array\":{\"type\":\"bytes\",\"data\":\"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\"},\"shape\":[4000],\"dtype\":\"float64\",\"order\":\"little\"}],[\"y\",{\"type\":\"ndarray\",\"array\":{\"type\":\"bytes\",\"data\":\"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\"},\"shape\":[4000],\"dtype\":\"float64\",\"order\":\"little\"}]]}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1346\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1347\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1342\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#1f77b3\",\"line_width\":2}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1343\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#1f77b3\",\"line_alpha\":0.1,\"line_width\":2}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Line\",\"id\":\"p1344\",\"attributes\":{\"x\":{\"type\":\"field\",\"field\":\"x\"},\"y\":{\"type\":\"field\",\"field\":\"y\"},\"line_color\":\"#1f77b3\",\"line_alpha\":0.2,\"line_width\":2}}}},{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1354\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1348\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1349\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1350\"},\"data\":{\"type\":\"map\"}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1355\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1356\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1351\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":0.4000369912873098},\"y\":{\"type\":\"value\",\"value\":0},\"angle\":{\"type\":\"value\",\"value\":3.141592653589793},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1352\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":0.4000369912873098},\"y\":{\"type\":\"value\",\"value\":0},\"angle\":{\"type\":\"value\",\"value\":3.141592653589793},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.1},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1353\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":0.4000369912873098},\"y\":{\"type\":\"value\",\"value\":0},\"angle\":{\"type\":\"value\",\"value\":3.141592653589793},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.2},\"line_width\":{\"type\":\"value\",\"value\":2}}}}},{\"type\":\"object\",\"name\":\"GlyphRenderer\",\"id\":\"p1363\",\"attributes\":{\"data_source\":{\"type\":\"object\",\"name\":\"ColumnDataSource\",\"id\":\"p1357\",\"attributes\":{\"selected\":{\"type\":\"object\",\"name\":\"Selection\",\"id\":\"p1358\",\"attributes\":{\"indices\":[],\"line_indices\":[]}},\"selection_policy\":{\"type\":\"object\",\"name\":\"UnionRenderers\",\"id\":\"p1359\"},\"data\":{\"type\":\"map\"}}},\"view\":{\"type\":\"object\",\"name\":\"CDSView\",\"id\":\"p1364\",\"attributes\":{\"filter\":{\"type\":\"object\",\"name\":\"AllIndices\",\"id\":\"p1365\"}}},\"glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1360\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":0.7198330350586437},\"y\":{\"type\":\"value\",\"value\":1},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"nonselection_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1361\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":0.7198330350586437},\"y\":{\"type\":\"value\",\"value\":1},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.1},\"line_width\":{\"type\":\"value\",\"value\":2}}},\"muted_glyph\":{\"type\":\"object\",\"name\":\"Ray\",\"id\":\"p1362\",\"attributes\":{\"x\":{\"type\":\"value\",\"value\":0.7198330350586437},\"y\":{\"type\":\"value\",\"value\":1},\"line_color\":{\"type\":\"value\",\"value\":\"#1f77b3\"},\"line_alpha\":{\"type\":\"value\",\"value\":0.2},\"line_width\":{\"type\":\"value\",\"value\":2}}}}}],\"toolbar\":{\"type\":\"object\",\"name\":\"Toolbar\",\"id\":\"p1315\",\"attributes\":{\"tools\":[{\"type\":\"object\",\"name\":\"PanTool\",\"id\":\"p1328\"},{\"type\":\"object\",\"name\":\"WheelZoomTool\",\"id\":\"p1329\",\"attributes\":{\"renderers\":\"auto\"}},{\"type\":\"object\",\"name\":\"BoxZoomTool\",\"id\":\"p1330\",\"attributes\":{\"overlay\":{\"type\":\"object\",\"name\":\"BoxAnnotation\",\"id\":\"p1331\",\"attributes\":{\"syncable\":false,\"level\":\"overlay\",\"visible\":false,\"left\":{\"type\":\"number\",\"value\":\"nan\"},\"right\":{\"type\":\"number\",\"value\":\"nan\"},\"top\":{\"type\":\"number\",\"value\":\"nan\"},\"bottom\":{\"type\":\"number\",\"value\":\"nan\"},\"left_units\":\"canvas\",\"right_units\":\"canvas\",\"top_units\":\"canvas\",\"bottom_units\":\"canvas\",\"line_color\":\"black\",\"line_alpha\":1.0,\"line_width\":2,\"line_dash\":[4,4],\"fill_color\":\"lightgrey\",\"fill_alpha\":0.5}}}},{\"type\":\"object\",\"name\":\"SaveTool\",\"id\":\"p1336\"},{\"type\":\"object\",\"name\":\"ResetTool\",\"id\":\"p1337\"},{\"type\":\"object\",\"name\":\"HelpTool\",\"id\":\"p1338\"}]}},\"toolbar_location\":\"above\",\"left\":[{\"type\":\"object\",\"name\":\"LinearAxis\",\"id\":\"p1323\",\"attributes\":{\"ticker\":{\"type\":\"object\",\"name\":\"BasicTicker\",\"id\":\"p1324\",\"attributes\":{\"mantissas\":[1,2,5]}},\"formatter\":{\"type\":\"object\",\"name\":\"BasicTickFormatter\",\"id\":\"p1325\"},\"axis_label\":\"ECDF\",\"major_label_policy\":{\"type\":\"object\",\"name\":\"AllLabels\",\"id\":\"p1326\"}}}],\"below\":[{\"type\":\"object\",\"name\":\"LinearAxis\",\"id\":\"p1318\",\"attributes\":{\"ticker\":{\"type\":\"object\",\"name\":\"BasicTicker\",\"id\":\"p1319\",\"attributes\":{\"mantissas\":[1,2,5]}},\"formatter\":{\"type\":\"object\",\"name\":\"BasicTickFormatter\",\"id\":\"p1320\"},\"axis_label\":\"std beak depth (mm)\",\"major_label_policy\":{\"type\":\"object\",\"name\":\"AllLabels\",\"id\":\"p1321\"}}}],\"center\":[{\"type\":\"object\",\"name\":\"Grid\",\"id\":\"p1322\",\"attributes\":{\"axis\":{\"id\":\"p1318\"}}},{\"type\":\"object\",\"name\":\"Grid\",\"id\":\"p1327\",\"attributes\":{\"dimension\":1,\"axis\":{\"id\":\"p1323\"}}}],\"frame_width\":375,\"frame_height\":275}}]}};\n", " const render_items = [{\"docid\":\"0b2e7d1b-00a1-4956-8235-e68a9cd87b0a\",\"roots\":{\"p1307\":\"d79a126b-db37-4388-b2ef-1e802f75de63\"},\"root_ids\":[\"p1307\"]}];\n", " void root.Bokeh.embed.embed_items_notebook(docs_json, render_items);\n", " }\n", " if (root.Bokeh !== undefined) {\n", " embed_document(root);\n", " } else {\n", " let attempts = 0;\n", " const timer = setInterval(function(root) {\n", " if (root.Bokeh !== undefined) {\n", " clearInterval(timer);\n", " embed_document(root);\n", " } else {\n", " attempts++;\n", " if (attempts > 100) {\n", " clearInterval(timer);\n", " console.log(\"Bokeh: ERROR: Unable to run BokehJS code because BokehJS library is missing\");\n", " }\n", " }\n", " }, 10, root)\n", " }\n", "})(window);" ], "application/vnd.bokehjs_exec.v0+json": "" }, "metadata": { "application/vnd.bokehjs_exec.v0+json": { "id": "p1307" } }, "output_type": "display_data" } ], "source": [ "# Compute replicates\n", "bs_reps_1975 = np.array([draw_bs_rep(bd_1975, np.std, rng) for _ in range(n_reps)])\n", " \n", "# Compute confidence interval\n", "conf_int_1975 = np.percentile(bs_reps_1975, [2.5, 97.5])\n", "print(conf_int_1975)\n", "\n", "# Plot ECDF\n", "p = iqplot.ecdf(\n", " bs_reps_1975,\n", " x_axis_label='std beak depth (mm)',\n", ")\n", "\n", "bokeh.io.show(p)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Note that the distribution for the standard deviation is not Normal; it has a heavier right tail. So, we now also have an estimate for the variability in beak depth. It could range from about 0.48 to 0.64 mm. We could report it like $0.56_{-0.08}^{+0.07}$ mm." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Computing environment" ] }, { "cell_type": "code", "execution_count": 20, "metadata": { "tags": [ "hide-input" ] }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Python implementation: CPython\n", "Python version : 3.11.9\n", "IPython version : 8.20.0\n", "\n", "numpy : 1.26.4\n", "pandas : 2.2.1\n", "bokeh : 3.4.0\n", "iqplot : 0.3.6\n", "jupyterlab: 4.0.13\n", "\n" ] } ], "source": [ "%load_ext watermark\n", "%watermark -v -p numpy,pandas,bokeh,iqplot,jupyterlab" ] } ], "metadata": { "anaconda-cloud": {}, "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": "3.11.9" } }, "nbformat": 4, "nbformat_minor": 4 }