{ "cells": [ { "cell_type": "markdown", "id": "eb81ed47", "metadata": {}, "source": [ "# Example 3: Bootstrap intuition - Mean vs Median\n", "\n", "**Students tasks**\n", "\n", "## Task A : Descriptive stats (Population=500,000)\n", "\n", "- [ ] Compute and report: mean, median, SD, IQR, five-number\n", "- [ ] Plot (histogram and boxplot)\n", "- [ ] Identify outlier using the 1.5IQR rule\n", "\n", "## Task B: Sampling variability\n", "\n", "- [ ] using the provided population, run repeated samplings for n=30, 50,100\n", "- [ ] Plot the sampling distribution (3 histograms)\n", "- [ ] Make a table with the empirical standard error (SD) vs n size\n", "\n", "## Task C: Bootstrap mean vs median\n", "\n", "- [ ] Bootstrap (resample with replacement from data) the mean and median (B=5000) for dataset with outliers (*)\n", "- [ ] Plot both bootstrap distributions\n", "- [ ] Compute 95% percentils CIs for the mean and median\n", "- [ ] Repeat after removing the the outliers and compare:\n", " - [ ] Which statistics changes more (mean or median)?\n", " - [ ] Which bootstrap distribution is wider, and why?" ] }, { "cell_type": "markdown", "id": "6abe3332", "metadata": {}, "source": [ "- [ ] Texto" ] } ], "metadata": { "language_info": { "name": "python" } }, "nbformat": 4, "nbformat_minor": 5 }