{
 "cells": [
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": [
    "# Area-to-area Poisson Kriging\n",
    "\n",
    "Before starting this tutorial, be sure that you have understood concepts in of ordinary kriging and semivariogram regularization.\n",
    "\n",
    "The Poisson Kriging technique is used to model spatial count data. We analyze a particular case where values are counted over blocks, and those polygons have irregular shapes and sizes. We will predict the breast cancer rates in the Northeastern counties of the U.S. We use U.S. Census 2010 data to create point support for blocks.\n",
    "\n",
    "In the previous tutorial `Poisson Kriging - Centroid based approach`, we've assumed that all areas have the same size and shape. It is not a valid statement. We now use the Area-to-Area Kriging technique to predict and evaluate missing values of counties with varying shapes and sizes.\n",
    "\n",
    "Area-to-Area and Area-to-Point Poisson Kriging are slower than simplified Kriging with Centroids but give more reliable results because they are tuned to the areal size and shape.\n",
    "\n",
    "## Prerequisites\n",
    "\n",
    "- **Domain**:\n",
    "  - kriging\n",
    "  - Poisson kriging and Poisson distribution\n",
    "  - rates\n",
    "\n",
    "- **Package**:\n",
    "  - `Deconvolution()`\n",
    "  - `ordinary_kriging()`\n",
    "  - `TheoreticalVariogram()`\n",
    "  - `centroid_poisson_kriging()`\n",
    "\n",
    "- **Programming**:\n",
    "  - Python basics\n",
    "  - `pandas` basics\n",
    "\n",
    "## Table of contents\n",
    "\n",
    "1. Prepare data\n",
    "2. Load regularized semivariogram model\n",
    "3. Prepare data for Poisson Kriging\n",
    "4. Filtering areas\n",
    "5. Evaluate\n",
    "\n",
    "## 1. Prepare data"
   ],
   "id": "86aebe1aafded181"
  },
  {
   "metadata": {
    "collapsed": true,
    "ExecuteTime": {
     "end_time": "2025-12-21T15:11:51.088926Z",
     "start_time": "2025-12-21T15:11:48.286540Z"
    }
   },
   "cell_type": "code",
   "source": [
    "import geopandas as gpd\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "from tqdm import tqdm\n",
    "\n",
    "from pyinterpolate import TheoreticalVariogram, Blocks, PointSupport, area_to_area_pk"
   ],
   "id": "initial_id",
   "outputs": [],
   "execution_count": 1
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-12-21T15:11:51.245010Z",
     "start_time": "2025-12-21T15:11:51.207493Z"
    }
   },
   "cell_type": "code",
   "source": [
    "DATASET = '../data/blocks/cancer_data.gpkg'\n",
    "OUTPUT = '../data/regularized_variogram.json'\n",
    "POLYGON_LAYER = 'areas'\n",
    "POPULATION_LAYER = 'points'\n",
    "POP10 = 'POP10'\n",
    "GEOMETRY_COL = 'geometry'\n",
    "POLYGON_ID = 'FIPS'\n",
    "POLYGON_VALUE = 'rate'"
   ],
   "id": "588576329b95c209",
   "outputs": [],
   "execution_count": 2
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-12-21T15:11:51.507336Z",
     "start_time": "2025-12-21T15:11:51.281059Z"
    }
   },
   "cell_type": "code",
   "source": [
    "blocks = Blocks(\n",
    "    ds=gpd.read_file(DATASET, layer=POLYGON_LAYER),\n",
    "    value_column_name=POLYGON_VALUE,\n",
    "    geometry_column_name=GEOMETRY_COL,\n",
    "    index_column_name=POLYGON_ID\n",
    ")"
   ],
   "id": "34b95d6b19d581cc",
   "outputs": [],
   "execution_count": 3
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-12-21T15:11:51.903394Z",
     "start_time": "2025-12-21T15:11:51.516906Z"
    }
   },
   "cell_type": "code",
   "source": [
    "point_support = PointSupport(\n",
    "    points=gpd.read_file(DATASET, layer=POPULATION_LAYER),\n",
    "    blocks=blocks,\n",
    "    points_value_column=POP10,\n",
    "    points_geometry_column=GEOMETRY_COL,\n",
    "    verbose=True\n",
    ")"
   ],
   "id": "ee50642839abfdd7",
   "outputs": [],
   "execution_count": 4
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-12-21T15:11:53.010076Z",
     "start_time": "2025-12-21T15:11:51.919408Z"
    }
   },
   "cell_type": "code",
   "source": [
    "blocks.ds.plot(column=blocks.value_column_name,\n",
    "               cmap='Spectral_r',\n",
    "               legend=True,\n",
    "               figsize=(12, 8))"
   ],
   "id": "a157978660c2ad0",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 1200x800 with 2 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 5
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": [
    "**Clarification**: It is always a good idea to look into the spatial patterns in dataset and to visually check if our data do not have any `NaN` values. We use the geopandas `GeoDataFrame.plot()` function with a color map that diverges regions based on the cancer incidence rates. The output map is not ideal, and it has a few unreliable results, for example:\n",
    "\n",
    "- In counties with a small population, where the ratio number of cases to population size is high even if the number of cases is low,\n",
    "- Counties that are very big and sparsely populated may draw more of our attention than densely populated counties,\n",
    "- Transitions of colors (rates) between counties may be too abrupt, even if we know that neighboring counties should have closer results.\n",
    "\n",
    "We can overcome those problems using Area-to-area Poisson Kriging for filtering and denoising.\n",
    "\n",
    "## 2. Load regularized semivariogram model\n",
    "\n",
    "We load a regularized semivariogram from the tutorial about Semivariogram Regularization. You can always perform semivariogram regularization along with the Poisson Kriging, but it is a very long process, and it is more convenient to separate those two steps."
   ],
   "id": "88d4c0129a969e1d"
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-12-21T15:11:53.296262Z",
     "start_time": "2025-12-21T15:11:53.101199Z"
    }
   },
   "cell_type": "code",
   "source": [
    "semivariogram = TheoreticalVariogram()\n",
    "semivariogram.from_json(OUTPUT)\n",
    "semivariogram.plot(experimental=False)"
   ],
   "id": "1956f4d48c8116f3",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 6
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": [
    "## 3. Prepare data for Poisson Kriging\n",
    "\n",
    "We simply select random blocks values from blocks data and try to predict those values using Poisson Kriging.\n"
   ],
   "id": "421b09804fee452b"
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-12-21T15:11:53.334026Z",
     "start_time": "2025-12-21T15:11:53.326405Z"
    }
   },
   "cell_type": "code",
   "source": [
    "sample_ids = blocks.ds.sample(20)['FIPS'].to_numpy()\n",
    "sample_ids"
   ],
   "id": "f617d5d25aa05105",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([42107, 36075, 42041, 42027, 34003, 36001, 42093, 42057,  9009,\n",
       "       25011, 23025, 36103, 34021, 42079, 36089, 36031, 33007, 42119,\n",
       "       36083, 23005])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 7
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": [
    "## 4. Filtering areas\n",
    "\n",
    "You may start to work with predictions with a fitted semivariogram model. Area-to-area Poisson Kriging takes five arguments during the run:\n",
    "\n",
    "1. **semivariogram_model**: fitted semivariogram model\n",
    "2. **point_support**: point support data\n",
    "3. **unknown_block_index**: id of the interpolated block (`point_support` must have this block's point support coordinates and values)\n",
    "4. **neighbors_range**: how wide is neighbors search area, by default it is `semivariogram_model` range\n",
    "5. **number_of_neighbors**: how many neighbors are used for the interpolation\n",
    "\n",
    "The algorithm in the cell below iteratively picks one area from the test set and performs prediction. Then, forecast bias, the difference between actual and predicted value, is calculated. Forecast Bias clarifies if our algorithm predictions are too low or too high (under- or overestimation). The following parameter is the Root Mean Squared Error. This measure tells us more about outliers and huge differences between predictions and actual values. We will see it in the last part of the tutorial.\n",
    "\n"
   ],
   "id": "bfa63ed6da0055b2"
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-12-21T15:11:57.820991Z",
     "start_time": "2025-12-21T15:11:53.360574Z"
    }
   },
   "cell_type": "code",
   "source": [
    "preds = []\n",
    "\n",
    "for b_id in tqdm(sample_ids):\n",
    "    pred = area_to_area_pk(\n",
    "        semivariogram_model=semivariogram,\n",
    "        point_support=point_support,\n",
    "        unknown_block_index=b_id,\n",
    "        neighbors_range=semivariogram.rang*2,\n",
    "        number_of_neighbors=8,\n",
    "        raise_when_negative_error=False\n",
    "    )\n",
    "    preds.append(pred)"
   ],
   "id": "c104ae2d0e3f593e",
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 20/20 [00:04<00:00,  4.53it/s]\n"
     ]
    }
   ],
   "execution_count": 8
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-12-21T15:11:57.835081Z",
     "start_time": "2025-12-21T15:11:57.831604Z"
    }
   },
   "cell_type": "code",
   "source": "results = pd.DataFrame(preds)",
   "id": "2ace5cb4b3c2d0f3",
   "outputs": [],
   "execution_count": 9
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-12-21T15:11:57.858156Z",
     "start_time": "2025-12-21T15:11:57.852090Z"
    }
   },
   "cell_type": "code",
   "source": "results = results.merge(blocks.ds, left_on='block_id', right_on='FIPS')",
   "id": "5849c2b0ea0e36f4",
   "outputs": [],
   "execution_count": 10
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-12-21T15:11:57.909432Z",
     "start_time": "2025-12-21T15:11:57.891424Z"
    }
   },
   "cell_type": "code",
   "source": "results.head()",
   "id": "5c70ac3406e6a97f",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "   block_id        zhat        sig   FIPS   rate  \\\n",
       "0     42107  128.049947   9.414107  42107  115.2   \n",
       "1     36075  132.715936  10.441738  36075  143.5   \n",
       "2     42041  119.579583   8.467769  42041  130.9   \n",
       "3     42027  118.067558  10.056972  42027  141.4   \n",
       "4     34003  127.193763   6.258667  34003  140.8   \n",
       "\n",
       "                                            geometry  \\\n",
       "0  MULTIPOLYGON (((1670088.257 358635.875, 167006...   \n",
       "1  MULTIPOLYGON (((1581876.402 689018.858, 158200...   \n",
       "2  MULTIPOLYGON (((1604893.567 299745.669, 160480...   \n",
       "3  MULTIPOLYGON (((1495040.309 391321.809, 149503...   \n",
       "4  MULTIPOLYGON (((1827117.746 423012.913, 182717...   \n",
       "\n",
       "            representative_points           lon            lat  \n",
       "0  POINT (1650323.464 368750.266)  1.650323e+06  368750.265783  \n",
       "1  POINT (1591216.884 663884.599)  1.591217e+06  663884.599117  \n",
       "2  POINT (1576325.693 291564.448)  1.576326e+06  291564.447617  \n",
       "3  POINT (1513734.416 363907.622)  1.513734e+06  363907.622101  \n",
       "4  POINT (1818622.729 437071.975)  1.818623e+06  437071.975383  "
      ],
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>block_id</th>\n",
       "      <th>zhat</th>\n",
       "      <th>sig</th>\n",
       "      <th>FIPS</th>\n",
       "      <th>rate</th>\n",
       "      <th>geometry</th>\n",
       "      <th>representative_points</th>\n",
       "      <th>lon</th>\n",
       "      <th>lat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>42107</td>\n",
       "      <td>128.049947</td>\n",
       "      <td>9.414107</td>\n",
       "      <td>42107</td>\n",
       "      <td>115.2</td>\n",
       "      <td>MULTIPOLYGON (((1670088.257 358635.875, 167006...</td>\n",
       "      <td>POINT (1650323.464 368750.266)</td>\n",
       "      <td>1.650323e+06</td>\n",
       "      <td>368750.265783</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>36075</td>\n",
       "      <td>132.715936</td>\n",
       "      <td>10.441738</td>\n",
       "      <td>36075</td>\n",
       "      <td>143.5</td>\n",
       "      <td>MULTIPOLYGON (((1581876.402 689018.858, 158200...</td>\n",
       "      <td>POINT (1591216.884 663884.599)</td>\n",
       "      <td>1.591217e+06</td>\n",
       "      <td>663884.599117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>42041</td>\n",
       "      <td>119.579583</td>\n",
       "      <td>8.467769</td>\n",
       "      <td>42041</td>\n",
       "      <td>130.9</td>\n",
       "      <td>MULTIPOLYGON (((1604893.567 299745.669, 160480...</td>\n",
       "      <td>POINT (1576325.693 291564.448)</td>\n",
       "      <td>1.576326e+06</td>\n",
       "      <td>291564.447617</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>42027</td>\n",
       "      <td>118.067558</td>\n",
       "      <td>10.056972</td>\n",
       "      <td>42027</td>\n",
       "      <td>141.4</td>\n",
       "      <td>MULTIPOLYGON (((1495040.309 391321.809, 149503...</td>\n",
       "      <td>POINT (1513734.416 363907.622)</td>\n",
       "      <td>1.513734e+06</td>\n",
       "      <td>363907.622101</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>34003</td>\n",
       "      <td>127.193763</td>\n",
       "      <td>6.258667</td>\n",
       "      <td>34003</td>\n",
       "      <td>140.8</td>\n",
       "      <td>MULTIPOLYGON (((1827117.746 423012.913, 182717...</td>\n",
       "      <td>POINT (1818622.729 437071.975)</td>\n",
       "      <td>1.818623e+06</td>\n",
       "      <td>437071.975383</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 11
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": [
    "## 5. Evaluate\n",
    "\n",
    "The analysis of errors is the last analysis step. We plot two histograms: forecast bias and squared error then we calculate the base statistics of a dataset."
   ],
   "id": "fc381cce8e6b6f71"
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-12-21T15:11:57.962285Z",
     "start_time": "2025-12-21T15:11:57.957103Z"
    }
   },
   "cell_type": "code",
   "source": [
    "results['forecast_bias'] = results['zhat'] - results['rate']\n",
    "results['squared_error'] = (results['rate'] - results['zhat'])**2"
   ],
   "id": "13dc5a512630e198",
   "outputs": [],
   "execution_count": 12
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-12-21T15:11:58.162280Z",
     "start_time": "2025-12-21T15:11:58.029316Z"
    }
   },
   "cell_type": "code",
   "source": "results['forecast_bias'].plot(kind='hist', bins=5)",
   "id": "c65a79959902a216",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: ylabel='Frequency'>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 13
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-12-21T15:11:58.383831Z",
     "start_time": "2025-12-21T15:11:58.186397Z"
    }
   },
   "cell_type": "code",
   "source": "results['squared_error'].plot(kind='hist', bins=5)",
   "id": "48b987512289d576",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: ylabel='Frequency'>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 14
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-12-21T15:11:58.427009Z",
     "start_time": "2025-12-21T15:11:58.404488Z"
    }
   },
   "cell_type": "code",
   "source": "results[['zhat', 'rate', 'forecast_bias', 'squared_error']].describe()",
   "id": "cecebe7de4cdb357",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "             zhat        rate  forecast_bias  squared_error\n",
       "count   20.000000   20.000000      20.000000      20.000000\n",
       "mean   129.228669  127.830000       1.398669     179.574443\n",
       "std      7.810987   14.454142      13.673569     215.335912\n",
       "min    117.906817  101.100000     -23.332442       0.008683\n",
       "25%    125.756270  115.575000      -8.189205       3.405850\n",
       "50%    127.729558  130.700000       0.657993      99.978315\n",
       "75%    131.912100  138.475000       9.628163     262.877367\n",
       "max    146.846616  152.900000      24.044155     578.121379"
      ],
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>zhat</th>\n",
       "      <th>rate</th>\n",
       "      <th>forecast_bias</th>\n",
       "      <th>squared_error</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>20.000000</td>\n",
       "      <td>20.000000</td>\n",
       "      <td>20.000000</td>\n",
       "      <td>20.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>129.228669</td>\n",
       "      <td>127.830000</td>\n",
       "      <td>1.398669</td>\n",
       "      <td>179.574443</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>7.810987</td>\n",
       "      <td>14.454142</td>\n",
       "      <td>13.673569</td>\n",
       "      <td>215.335912</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>117.906817</td>\n",
       "      <td>101.100000</td>\n",
       "      <td>-23.332442</td>\n",
       "      <td>0.008683</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>125.756270</td>\n",
       "      <td>115.575000</td>\n",
       "      <td>-8.189205</td>\n",
       "      <td>3.405850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>127.729558</td>\n",
       "      <td>130.700000</td>\n",
       "      <td>0.657993</td>\n",
       "      <td>99.978315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>131.912100</td>\n",
       "      <td>138.475000</td>\n",
       "      <td>9.628163</td>\n",
       "      <td>262.877367</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>146.846616</td>\n",
       "      <td>152.900000</td>\n",
       "      <td>24.044155</td>\n",
       "      <td>578.121379</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 15
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-12-21T15:11:58.493477Z",
     "start_time": "2025-12-21T15:11:58.489121Z"
    }
   },
   "cell_type": "code",
   "source": [
    "rmse = np.sqrt(np.mean(results['squared_error']))\n",
    "print(rmse)"
   ],
   "id": "d36c13300a44d17",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "13.400538906042224\n"
     ]
    }
   ],
   "execution_count": 16
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": [
    "**Clarification**: Analysis of Forecast Bias and Root Mean Squared Error. Their distribution and basic properties are describing model performance. However, consider that the table above is a single test case (realization) and can be misleading. The good idea is to repeat the test dozens of times with a different training/test set division each time. After this, we average results from multiple tests and get insight into our model's behavior.\n",
    "\n",
    "- negative forecast bias: model underestimates predictions\n",
    "- positive forecast bias: model overestimates predictions\n",
    "- forecast bias close to 0: model is (probably) accurate\n",
    "- root mean squared error: accuracy measure, large rmse means that model performs poorly or we have many outliers\n",
    "\n",
    "**Note 1**: Those results are not decisive. Our sample has been selected randomly, and there is a chance that it is not a spatially representative sample! (E.g., areas only from one region). The good idea is to repeat the experiment multiple times with other samples and average results to determine how well the model performs.\n",
    "\n",
    "**Note 2**: If we analyze errors’ statistics, we should consider not only error's mean value. Let’s look at different pieces of information:\n",
    "\n",
    "- Histograms clearly show us how dispersed and grouped errors are, and most importantly, we directly see the worst predictions and how many of them are generated by our model.\n",
    "- What is plotted on a histogram is described by statistics. We may check the max and the min error, but the true power comes when we analyze quartiles and standard deviation.\n",
    "- The standard deviation is a good measurement of our model’s variance; lower standard deviation of errors, the better.\n",
    "- Sometimes, we must look into quartiles. A very high mean but relatively low median (or even the 3rd quartile) indicates that we have only a few wrong predictions (that could be interesting, because it means that few locations are heavily under- or over-estimated).\n",
    "\n",
    "**Note 3**: Root Mean Squared Error from `Area-to-Area Poisson Kriging` should be lower than RMSE from `Centroid-based Poisson Kriging` because predictions are more smoothed by the fact that ATA takes into account population distribution within an area.\n"
   ],
   "id": "e8f72501fe225c1f"
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": [
    "## Changelog\n",
    "\n",
    "| Date       | Changes                                      | Author                           |\n",
    "|------------|----------------------------------------------|----------------------------------|\n",
    "| 2025-05-30 | Tutorial has been adapted to the 1.0 release | @SimonMolinsky (Szymon Moliński) |"
   ],
   "id": "46044ec1bb5dafc0"
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}