{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Working with Percentiles (and Quantiles) of Distributions\n",
    "\n",
    "Many times we want to observe different distributional properties of scenario data. The **pyam** function `compute.quantiles()` can help!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 0. Define timeseries data and initialize an IamDataFrame\n",
    "\n",
    "This tutorial uses a scenario similar to the data in the **first-steps tutorial** (here on\n",
    "[GitHub](https://github.com/IAMconsortium/pyam/blob/master/doc/source/tutorials/pyam_first_steps.ipynb)\n",
    "and on [read the docs](https://pyam-iamc.readthedocs.io/en/stable/tutorials/pyam_first_steps.html)).  \n",
    "Please read that tutorial for the reference and further information."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pyam import IamDataFrame\n",
    "\n",
    "df = IamDataFrame(data=\"tutorial_data.csv\")\n",
    "df.timeseries().head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Let's see how many scenarios define CO2 emissions\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.filter(variable=\"Emissions|CO2\", region=\"World\").scenario"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Get the median\n",
    "\n",
    "The median is the `0.5` quantile (or percentile) - let's take a look!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "\n",
    "# plot the background field of scenario data\n",
    "(\n",
    "    df.filter(variable=\"Emissions|CO2\", region=\"World\").plot.line(\n",
    "        color=\"variable\", alpha=0, fill_between=True, ax=ax\n",
    "    )\n",
    ")\n",
    "\n",
    "# plot just the median\n",
    "(\n",
    "    df.filter(variable=\"Emissions|CO2\", region=\"World\")\n",
    "    .compute.quantiles([0.5])\n",
    "    .plot.line(ax=ax)\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Get arbitrary quantiles\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots()\n",
    "\n",
    "# plot the background field of scenario data\n",
    "(\n",
    "    df.filter(variable=\"Emissions|CO2\", region=\"World\").plot.line(\n",
    "        color=\"variable\", alpha=0, fill_between=True, ax=ax\n",
    "    )\n",
    ")\n",
    "\n",
    "# plot quantiles\n",
    "(\n",
    "    df.filter(variable=\"Emissions|CO2\", region=\"World\")\n",
    "    .compute.quantiles([0.2, 0.6, 0.8])\n",
    "    .plot.line(ax=ax)\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4. Weighted Quantiles\n",
    "\n",
    "Weighted quantiles are also supported via the `wquantiles` package and are keyed to model/scenario combinations (unless the `level` argument is provided to `compute.quantiles()`). "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "weights = df.meta.assign(weight=np.random.rand(len(df.meta)))\n",
    "weights.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fig, ax = plt.subplots()\n",
    "\n",
    "# plot the background field of scenario data\n",
    "(\n",
    "    df.filter(variable=\"Emissions|CO2\", region=\"World\").plot.line(\n",
    "        color=\"variable\", alpha=0, fill_between=True, ax=ax\n",
    "    )\n",
    ")\n",
    "\n",
    "# plot weighted quantiles\n",
    "(\n",
    "    df.filter(variable=\"Emissions|CO2\", region=\"World\")\n",
    "    .compute.quantiles([0.2, 0.6, 0.8], weights=weights[\"weight\"])\n",
    "    .plot.line(ax=ax)\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.10.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}