{ "cells": [ { "cell_type": "markdown", "id": "bd107155-704e-43b9-adb6-cafe692d3124", "metadata": {}, "source": [ "(lecture_20)=\n", "# Horoscopes\n", ":::{post} Jan 7, 2024\n", ":tags: statistical rethinking, bayesian inference, scientific workflow\n", ":category: intermediate\n", ":author: Dustin Stansbury\n", ":::\n", "\n", "This notebook is part of the PyMC port of the [Statistical Rethinking 2023](https://github.com/rmcelreath/stat_rethinking_2023) lecture series by Richard McElreath.\n", "\n", "[Video - Lecture 20 - Horoscopes](https://youtu.be/qwF-st2NGTU)# [Lecture 20 - Horoscopes](https://www.youtube.com/watch?v=qwF-st2NGTU)" ] }, { "cell_type": "code", "execution_count": 1, "id": "f5da83c4-9f52-488f-8039-bfcaba2c63d6", "metadata": {}, "outputs": [], "source": [ "# Ignore warnings\n", "import warnings\n", "\n", "import arviz as az\n", "import numpy as np\n", "import pandas as pd\n", "import pymc as pm\n", "import statsmodels.formula.api as smf\n", "import utils as utils\n", "import xarray as xr\n", "\n", "from matplotlib import pyplot as plt\n", "from matplotlib import style\n", "from scipy import stats as stats\n", "\n", "warnings.filterwarnings(\"ignore\")\n", "\n", "# Set matplotlib style\n", "STYLE = \"statistical-rethinking-2023.mplstyle\"\n", "style.use(STYLE)" ] }, { "cell_type": "markdown", "id": "a904010e-add4-4540-a908-d841d1a7f728", "metadata": {}, "source": [ "# Horoscopes\n", "\n", "This lecture mostly outlines a set of high-level heuristics and workflows to improve the quality of scientific research. Therefore there's not a lot of implementation details in the lecture to cover. I won't go through copying the content from each slide, but I cover some highlights (mostly for my own benefit) below:\n", "\n", "## Statistics is like fortune telling\n", "\n", "- Vague facts lead to vague advice\n", " - Reading tea leaves is like following common flow charts for statistical analysis\n", " - There's little scientific inputs, therefore little scientific interpretation\n", " - That's often the feature and the bug of fortune telling, and statistics:\n", " - by providing vague interpretations (e.g. horoscope predictions) from vague inputs (e.g. birthday), they can \"explain\" any number of outcomes\n", " - just like vague horoscopes can \"explain\" any number of possible future events\n", "- Exaggerated importance\n", " - no one wants to hear evil portents in their tea leaves, just as no one wants to hear about NULL or negative statistical results\n", " - there's often incentive to use statistics to find the positive result\n", "- It's often easier to offload subjective scientific responsibility onto objective statistical procedures\n", "\n", "## Three pillars of scientific workflow\n", "**1. Planning**\n", " - Goal setting\n", " - estimands\n", " - Theory building\n", " - assumptions\n", " - 4 types of theory building, increasing in specificity\n", " 1. Heuristic (DAGs)\n", " - allows us to deduce a lot from establishing causal structure\n", " 3. Structural\n", " - moves beyond DAGs by establishing specific functional forms of causes\n", " 5. Dynamical models\n", " - usually work over spatial/temporal grid\n", " - tend to collapse large number of micro-states into macro interpretation\n", " 7. Agent-based\n", " - focuses on individual micro states\n", " - Justified sampling\n", " - Which data do we use, and what's it's structure\n", " - Verify with simulation\n", " - Justified analysis\n", " - Which golems?\n", " - Can we recover estimands from simulations?\n", " - Documentation\n", " - How did it happen?\n", " - Help others and your future self\n", " - Scripting is self-documenting\n", " - Comments are important\n", " - Don't be clever, be explicit\n", " - Avoid clever one-liners\n", " - I find Python PEP useful here\n", " - Sharing\n", " - open source code and data formats\n", " - proprietary software does not facilitate shareing, and is bad scientific ethics\n", " - the irony here, is that MATLAB is so common in academic setting, particularly engineering 🙄\n", " - proprietery data formats can shoot you in the foot when you (or others) can no longer open them\n", " - Preregistration isn't a silver bullet\n", " - Pre-allocating expectations on a bad analysis approach (e.g. causal salad) doesn't fix the bad approach\n", "\n", "**2. Working**\n", "\n", "- Research engineering\n", " - Treat research more like software enginnering\n", " - standardized, battle-tested procedures that make software dependable and repeatable\n", " - version control (git)\n", " - testing\n", " - unit testing\n", " - integration testing\n", " - build up tests incrementally, validating each part of the workflow before proceeding to the next\n", " - documentation\n", " - review\n", " - 👀, 👀 have at least one other person review your analysis code and docs, and provide feedback\n", " - will often point out bugs, optimizations, or shortcomings in documentation\n", "- Look at good examples\n", " - e.g. on of [McElreath's Consulting Projects](https://github.com/rmcelreath/CES_rater_2021/tree/main)\n", " - [Data Carpentry](https://datacarpentry.org/)\n", "\n", "**3. Reporting**\n", "\n", "- Sharing materials\n", " - by following code-based Working flow, sharing is pretty much done for you\n", " - [Nice breakdown and example of Describing Methods](https://youtu.be/qwF-st2NGTU?si=3I7CMalLXv3pIQhr&t=3742)\n", "- Justify priors\n", "- Justify methods, and dealing with reviewers\n", " - Common fallacy: \"good scientific design doesn't require complex statistics\"\n", " - valid causal modeling requires complexity\n", " - don't try to convince Reviewer 3 to accept your methods, write to editor\n", " - move the convo from statistical to causal modeling\n", "- Describe data\n", " - structure\n", " - missing values: justify imputation if any\n", "- Describe results\n", " - aim to report contrasts and marginal effects\n", " - use densities over intervals\n", " - avoid interpeting coefficients as causal effects\n", "- Making decisions\n", " - this is often the goal (particularly in industry)\n", " - embrace uncertainty\n", " - uncertainty is not admission of weakness\n", " - Bayesian decision theory\n", " - use the posterior to simulate various policy interventions\n", " - can be used to provide posteriors to costs/benefits due to those interventions" ] }, { "cell_type": "markdown", "id": "e107b583-64df-4bf2-96a8-5a1e0e5c2e44", "metadata": {}, "source": [ "## Scientific Reform\n", "\n", "- many of the metrics for good science are counterproductive\n", " - e.g. papers that are least replicated continue to have higher citation count\n", " - META POINT: this result in publishing be explained using a causal modeling and colider bias\n", " \n", "### Collider bias in scientific publishing\n", "\n", "#### Causal model of collider bias" ] }, { "cell_type": "code", "execution_count": 2, "id": "a8f37da9-9b26-4a76-ba12-7d078f5e9257", "metadata": { "jupyter": { "source_hidden": true } }, "outputs": [ { "data": { "image/svg+xml": [ "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "\n", "newsworhiness, N\n", "\n", "newsworhiness, N\n", "\n", "\n", "\n", "published, P\n", "\n", "published, P\n", "\n", "\n", "\n", "newsworhiness, N->published, P\n", "\n", "\n", "\n", "\n", "\n", "trustworthiness, T\n", "\n", "trustworthiness, T\n", "\n", "\n", "\n", "trustworthiness, T->published, P\n", "\n", "\n", "\n", "\n", "\n" ], "text/plain": [ "" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "utils.draw_causal_graph(\n", " edge_list=[(\"newsworhiness, N\", \"published, P\"), (\"trustworthiness, T\", \"published, P\")],\n", ")" ] }, { "cell_type": "markdown", "id": "a80640fd-5392-4c2e-b567-effdc241dbb3", "metadata": {}, "source": [ "#### Simulating data from collider causal model" ] }, { "cell_type": "code", "execution_count": 3, "id": "4e4e1df3-5d3f-47f3-b247-004d57637346", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "np.random.seed(123)\n", "n_samples = 200\n", "\n", "# N and T are independent\n", "N = stats.norm.rvs(size=n_samples)\n", "T = stats.norm.rvs(size=n_samples)\n", "\n", "# Award criterion; either are large enough to threshold\n", "A = np.where(N + T > 2, 1, 0)\n", "\n", "for awarded in [0, 1]:\n", " color = \"gray\" if not awarded else \"C0\"\n", " N_A = N[A == awarded]\n", " T_A = T[A == awarded]\n", " utils.plot_scatter(N_A, T_A, color=color)\n", "\n", "fit_data = pd.DataFrame({\"N\": N_A, \"T\": T_A})\n", "cc = np.corrcoef(fit_data.T, fit_data.N)[0][1]\n", "awarded_model = smf.ols(\"T ~ N\", data=fit_data).fit()\n", "utils.plot_line(\n", " N_A, awarded_model.predict(), color=\"C0\", label=f\"Post-selection\\nTrend\\ncorrelation={cc:0.2}\"\n", ")\n", "plt.xlabel(\"Newsworthiness\")\n", "plt.ylabel(\"Trustworthiness\")\n", "plt.axis(\"square\")\n", "plt.legend();" ] }, { "cell_type": "markdown", "id": "e9ca00eb-a348-4ef6-9979-f5727663282c", "metadata": {}, "source": [ "By selecting at papers that are published based on a threshold that combines either newsworthiness--i.e. \"sexy papers\" that get cited a lot--or trustworthiness--i.e. boring papers that are replicable--we end up with highly-cited papers that tend to be less replicable." ] }, { "cell_type": "markdown", "id": "c361ac2b-1a9d-4961-af81-2ab1a3e47170", "metadata": { "jp-MarkdownHeadingCollapsed": true }, "source": [ "## Horoscopes of research\n", "\n", "- Many things that are \"bad\" about science (e.g. impact factor) are once well-intentioned reforms\n", "- Some potential fixes are available:\n", " 1. No stats before transparently-communicated causal model\n", " - avoid causal salad\n", " 2. Prove your code/analysis works within the scope of your project and assumptions\n", " 3. Share as much as possible\n", " - sometimes data is not shareable\n", " - but you can create partial, anonomized, or synthetic datasets\n", " 4. Beware proxies for research quality (e.g. citation count, impact factor)" ] }, { "cell_type": "markdown", "id": "007ca495", "metadata": {}, "source": [ "## Authors\n", "* Ported to PyMC by Dustin Stansbury (2024)\n", "* Based on Statistical Rethinking (2023) lectures by Richard McElreath" ] }, { "cell_type": "code", "execution_count": 4, "id": "2e6d8594", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Last updated: Tue Dec 17 2024\n", "\n", "Python implementation: CPython\n", "Python version : 3.12.5\n", "IPython version : 8.27.0\n", "\n", "pytensor: 2.26.4\n", "xarray : 2024.7.0\n", "\n", "scipy : 1.14.1\n", "xarray : 2024.7.0\n", "pymc : 5.19.1\n", "numpy : 1.26.4\n", "matplotlib : 3.9.2\n", "statsmodels: 0.14.2\n", "arviz : 0.19.0\n", "pandas : 2.2.2\n", "\n", "Watermark: 2.5.0\n", "\n" ] } ], "source": [ "%load_ext watermark\n", "%watermark -n -u -v -iv -w -p pytensor,xarray" ] }, { "cell_type": "markdown", "id": "494b5652", "metadata": {}, "source": [ ":::{include} ../page_footer.md\n", ":::" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.12.5" } }, "nbformat": 4, "nbformat_minor": 5 }