diff --git a/session-2-intuition/bootstrap.ipynb b/session-2-intuition/bootstrap.ipynb new file mode 100644 index 0000000..ebd4daf --- /dev/null +++ b/session-2-intuition/bootstrap.ipynb @@ -0,0 +1,50 @@ +{ + "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\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 +}