From 21c5b242501d020f6e93bac0352c138b6313e50f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Micha=C5=82=20Sar?=
 <101715585+michal-sar@users.noreply.github.com>
Date: Mon, 6 May 2024 22:55:21 +0200
Subject: [PATCH] move project

---
 main.ipynb | 2191 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 2191 insertions(+)
 create mode 100644 main.ipynb

diff --git a/main.ipynb b/main.ipynb
new file mode 100644
index 00000000..86456ff7
--- /dev/null
+++ b/main.ipynb
@@ -0,0 +1,2191 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "..."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "pd.options.display.float_format = '{:.2f}'.format"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "..."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "C:\\Users\\micha\\AppData\\Local\\Temp\\ipykernel_26160\\3760256257.py:1: DtypeWarning: Columns (25) have mixed types. Specify dtype option on import or set low_memory=False.\n",
+      "  df_dofinansowanie = pd.read_csv(\n"
+     ]
+    }
+   ],
+   "source": [
+    "df_dofinansowanie = pd.read_csv(\n",
+    "    'umowy_pelna_lista_krajowe.csv',\n",
+    "    encoding='ISO-8859-2',\n",
+    "    converters={'TERYT pe?ny': str},\n",
+    "    thousands=',')\n",
+    "\n",
+    "df_dofinansowanie = df_dofinansowanie.loc[df_dofinansowanie['TERYT pe?ny'] != ''].reset_index(drop=True)\n",
+    "\n",
+    "df_dofinansowanie['Dofinansowanie UE (PLN)'] = \\\n",
+    "    df_dofinansowanie['Dofinansowanie UE (PLN)'].apply(pd.to_numeric)\n",
+    "\n",
+    "df_dofinansowanie['Data rozpocz?cia realizacji'] = pd.to_datetime(df_dofinansowanie['Data rozpocz?cia realizacji'])\n",
+    "df_dofinansowanie['Rok rozpocz?cia realizacji'] = df_dofinansowanie['Data rozpocz?cia realizacji'].dt.year\n",
+    "\n",
+    "df_dofinansowanie['Data podpisania umowy pierwotnej'] = pd.to_datetime(df_dofinansowanie['Data podpisania umowy pierwotnej'])\n",
+    "df_dofinansowanie['Rok podpisania umowy pierwotnej'] = df_dofinansowanie['Data podpisania umowy pierwotnej'].dt.year"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df_dofinansowanie_agg = df_dofinansowanie \\\n",
+    "    .groupby(['TERYT pe?ny', 'Rok rozpocz?cia realizacji'])['Dofinansowanie UE (PLN)'].sum().reset_index()\n",
+    "df_dofinansowanie_agg = df_dofinansowanie_agg \\\n",
+    "    .rename(columns={'TERYT pe?ny': 'Kod', 'Rok rozpocz?cia realizacji': 'Rok', 'Dofinansowanie UE (PLN)': 'Suma'})\n",
+    "df_dofinansowanie_agg = df_dofinansowanie_agg \\\n",
+    "    .loc[df_dofinansowanie_agg['Kod'].str.len() == 7].reset_index(drop=True)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "..."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df_podz = pd.read_csv(\n",
+    "    'PODZ_1410_CREL.csv',\n",
+    "    sep=';',\n",
+    "    converters={'Kod': str})\n",
+    "df_podz = df_podz[['Kod', 'Rok', 'Wartosc']]\n",
+    "df_podz = df_podz.loc[df_podz['Kod'].str.endswith(('1', '2', '3'))]\n",
+    "df_podz = df_podz.dropna()\n",
+    "df_podz = df_podz.rename(columns={\n",
+    "    'Wartosc': 'Powierzchnia'})"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df_wyna = pd.read_csv(\n",
+    "    'WYNA_2497_CREL.csv',\n",
+    "    sep=';',\n",
+    "    converters={'Kod': str},\n",
+    "    decimal=',')\n",
+    "df_wyna = df_wyna[['Kod', 'Wyszczególnienie', 'Rok', 'Wartosc']]\n",
+    "df_wyna = df_wyna.dropna()\n",
+    "df_wyna = df_wyna.pivot_table(index=['Kod', 'Rok'], columns='Wyszczególnienie', values='Wartosc').reset_index()\n",
+    "df_wyna = df_wyna.rename(columns={\n",
+    "    'ogółem': 'Wynagrodzenie_ogolem',\n",
+    "    'przeciętne miesięczne wynagrodzenia brutto w relacji do średniej krajowej (Polska=100)': 'Wynagrodzenie_w_relacji_do_sredniej'})"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "C:\\Users\\micha\\AppData\\Local\\Temp\\ipykernel_26160\\3644708252.py:1: DtypeWarning: Columns (7) have mixed types. Specify dtype option on import or set low_memory=False.\n",
+      "  df_fina_1 = pd.read_csv(\n"
+     ]
+    },
+    {
+     "data": {
+      "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>Rodzaje dochodów</th>\n",
+       "      <th>Kod</th>\n",
+       "      <th>Rok</th>\n",
+       "      <th>Dochody_podatek_lesny</th>\n",
+       "      <th>Dochody_podatek_PCC</th>\n",
+       "      <th>Dochody_podatek_...</th>\n",
+       "      <th>Dochody_podatek_od_nieruchomosci</th>\n",
+       "      <th>Dochody_podatek_od_spadkow</th>\n",
+       "      <th>Dochody_podatek_od_srodkow_transportowych</th>\n",
+       "      <th>Dochody_podatek_rolny</th>\n",
+       "      <th>Dochody_podatek_odrebne_ustawy</th>\n",
+       "      <th>Dochody_razem</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2004</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>549608.00</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>13532989.00</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>625159.00</td>\n",
+       "      <td>23687.00</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>41378568.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2005</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>609855.00</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>13667398.00</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>700134.00</td>\n",
+       "      <td>26634.00</td>\n",
+       "      <td>15438121.00</td>\n",
+       "      <td>43417443.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2006</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>844223.65</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>14633962.72</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>747182.64</td>\n",
+       "      <td>11683.60</td>\n",
+       "      <td>16647124.98</td>\n",
+       "      <td>50319253.08</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2007</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>1344365.01</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>14944781.74</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>777345.52</td>\n",
+       "      <td>19377.36</td>\n",
+       "      <td>17436387.93</td>\n",
+       "      <td>62025513.24</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2008</td>\n",
+       "      <td>6799.55</td>\n",
+       "      <td>1790135.40</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>16089534.56</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>836441.10</td>\n",
+       "      <td>30823.60</td>\n",
+       "      <td>19149551.45</td>\n",
+       "      <td>80755930.93</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>47078</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2018</td>\n",
+       "      <td>154462.39</td>\n",
+       "      <td>5361951.37</td>\n",
+       "      <td>572868.36</td>\n",
+       "      <td>108107448.79</td>\n",
+       "      <td>437144.83</td>\n",
+       "      <td>589658.88</td>\n",
+       "      <td>51297.75</td>\n",
+       "      <td>115274832.37</td>\n",
+       "      <td>261780766.79</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>47079</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2019</td>\n",
+       "      <td>150329.31</td>\n",
+       "      <td>6088184.20</td>\n",
+       "      <td>468411.51</td>\n",
+       "      <td>38527846.59</td>\n",
+       "      <td>228886.23</td>\n",
+       "      <td>608637.40</td>\n",
+       "      <td>64855.15</td>\n",
+       "      <td>46137150.39</td>\n",
+       "      <td>167638796.15</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>47080</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2020</td>\n",
+       "      <td>156556.52</td>\n",
+       "      <td>5125090.74</td>\n",
+       "      <td>329522.12</td>\n",
+       "      <td>78767466.83</td>\n",
+       "      <td>552009.16</td>\n",
+       "      <td>558925.68</td>\n",
+       "      <td>48689.09</td>\n",
+       "      <td>85538260.14</td>\n",
+       "      <td>263006955.07</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>47081</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2021</td>\n",
+       "      <td>163778.36</td>\n",
+       "      <td>9082482.28</td>\n",
+       "      <td>492045.28</td>\n",
+       "      <td>78491368.35</td>\n",
+       "      <td>947992.83</td>\n",
+       "      <td>602586.14</td>\n",
+       "      <td>59824.46</td>\n",
+       "      <td>89840077.70</td>\n",
+       "      <td>252345800.93</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>47082</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2022</td>\n",
+       "      <td>174823.49</td>\n",
+       "      <td>7474079.65</td>\n",
+       "      <td>1019054.56</td>\n",
+       "      <td>84996948.99</td>\n",
+       "      <td>593315.54</td>\n",
+       "      <td>627169.86</td>\n",
+       "      <td>50987.00</td>\n",
+       "      <td>94936379.09</td>\n",
+       "      <td>259310641.60</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>47083 rows × 11 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "Rodzaje dochodów      Kod   Rok  Dochody_podatek_lesny  Dochody_podatek_PCC   \n",
+       "0                 0201011  2004                    NaN            549608.00  \\\n",
+       "1                 0201011  2005                    NaN            609855.00   \n",
+       "2                 0201011  2006                    NaN            844223.65   \n",
+       "3                 0201011  2007                    NaN           1344365.01   \n",
+       "4                 0201011  2008                6799.55           1790135.40   \n",
+       "...                   ...   ...                    ...                  ...   \n",
+       "47078             3263011  2018              154462.39           5361951.37   \n",
+       "47079             3263011  2019              150329.31           6088184.20   \n",
+       "47080             3263011  2020              156556.52           5125090.74   \n",
+       "47081             3263011  2021              163778.36           9082482.28   \n",
+       "47082             3263011  2022              174823.49           7474079.65   \n",
+       "\n",
+       "Rodzaje dochodów  Dochody_podatek_...  Dochody_podatek_od_nieruchomosci   \n",
+       "0                                 NaN                       13532989.00  \\\n",
+       "1                                 NaN                       13667398.00   \n",
+       "2                                 NaN                       14633962.72   \n",
+       "3                                 NaN                       14944781.74   \n",
+       "4                                 NaN                       16089534.56   \n",
+       "...                               ...                               ...   \n",
+       "47078                       572868.36                      108107448.79   \n",
+       "47079                       468411.51                       38527846.59   \n",
+       "47080                       329522.12                       78767466.83   \n",
+       "47081                       492045.28                       78491368.35   \n",
+       "47082                      1019054.56                       84996948.99   \n",
+       "\n",
+       "Rodzaje dochodów  Dochody_podatek_od_spadkow   \n",
+       "0                                        NaN  \\\n",
+       "1                                        NaN   \n",
+       "2                                        NaN   \n",
+       "3                                        NaN   \n",
+       "4                                        NaN   \n",
+       "...                                      ...   \n",
+       "47078                              437144.83   \n",
+       "47079                              228886.23   \n",
+       "47080                              552009.16   \n",
+       "47081                              947992.83   \n",
+       "47082                              593315.54   \n",
+       "\n",
+       "Rodzaje dochodów  Dochody_podatek_od_srodkow_transportowych   \n",
+       "0                                                 625159.00  \\\n",
+       "1                                                 700134.00   \n",
+       "2                                                 747182.64   \n",
+       "3                                                 777345.52   \n",
+       "4                                                 836441.10   \n",
+       "...                                                     ...   \n",
+       "47078                                             589658.88   \n",
+       "47079                                             608637.40   \n",
+       "47080                                             558925.68   \n",
+       "47081                                             602586.14   \n",
+       "47082                                             627169.86   \n",
+       "\n",
+       "Rodzaje dochodów  Dochody_podatek_rolny  Dochody_podatek_odrebne_ustawy   \n",
+       "0                              23687.00                             NaN  \\\n",
+       "1                              26634.00                     15438121.00   \n",
+       "2                              11683.60                     16647124.98   \n",
+       "3                              19377.36                     17436387.93   \n",
+       "4                              30823.60                     19149551.45   \n",
+       "...                                 ...                             ...   \n",
+       "47078                          51297.75                    115274832.37   \n",
+       "47079                          64855.15                     46137150.39   \n",
+       "47080                          48689.09                     85538260.14   \n",
+       "47081                          59824.46                     89840077.70   \n",
+       "47082                          50987.00                     94936379.09   \n",
+       "\n",
+       "Rodzaje dochodów  Dochody_razem  \n",
+       "0                   41378568.00  \n",
+       "1                   43417443.00  \n",
+       "2                   50319253.08  \n",
+       "3                   62025513.24  \n",
+       "4                   80755930.93  \n",
+       "...                         ...  \n",
+       "47078              261780766.79  \n",
+       "47079              167638796.15  \n",
+       "47080              263006955.07  \n",
+       "47081              252345800.93  \n",
+       "47082              259310641.60  \n",
+       "\n",
+       "[47083 rows x 11 columns]"
+      ]
+     },
+     "execution_count": 27,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df_fina_1 = pd.read_csv(\n",
+    "    'FINA_2622_CREL_1.csv',\n",
+    "    sep=';',\n",
+    "    converters={'Kod': str},\n",
+    "    decimal=',')\n",
+    "df_fina_1 = df_fina_1[['Kod', 'Rodzaje dochodów', 'Rok', 'Wartosc']]\n",
+    "df_fina_1 = df_fina_1.dropna()\n",
+    "df_fina_1 = df_fina_1.pivot_table(index=['Kod', 'Rok'], columns='Rodzaje dochodów', values='Wartosc').reset_index()\n",
+    "df_fina_1 = df_fina_1.rename(columns={\n",
+    "    'dochody podatkowe - podatek leśny': 'Dochody_podatek_lesny',\n",
+    "    'dochody podatkowe - podatek od czynności cywilnoprawnych': 'Dochody_podatek_PCC',\n",
+    "    'dochody podatkowe - podatek od działalności gospodarczej osób fizycznych, opłacany w formie karty podatkowej': 'Dochody_podatek_...',\n",
+    "    'dochody podatkowe - podatek od nieruchomości': 'Dochody_podatek_od_nieruchomosci',\n",
+    "    'dochody podatkowe - podatek od spadków i darowizn': 'Dochody_podatek_od_spadkow',\n",
+    "    'dochody podatkowe - podatek od środków transportowych': 'Dochody_podatek_od_srodkow_transportowych',\n",
+    "    'dochody podatkowe - podatek rolny': 'Dochody_podatek_rolny',\n",
+    "    'dochody podatkowe - ustalone i pobierane na podstawie odrębnych ustaw': 'Dochody_podatek_odrebne_ustawy',\n",
+    "    'razem': 'Dochody_razem'})\n",
+    "\n",
+    "df_fina_1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 28,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "C:\\Users\\micha\\AppData\\Local\\Temp\\ipykernel_26160\\2161929356.py:1: DtypeWarning: Columns (7) have mixed types. Specify dtype option on import or set low_memory=False.\n",
+      "  df_fina_2 = pd.read_csv(\n"
+     ]
+    },
+    {
+     "data": {
+      "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>Rodzaje dochodów</th>\n",
+       "      <th>Kod</th>\n",
+       "      <th>Rok</th>\n",
+       "      <th>Dochody_z_majatku</th>\n",
+       "      <th>Dochody_z_najmu_i_dzierzawy</th>\n",
+       "      <th>Dochody_z_uslug</th>\n",
+       "      <th>Dochody_dofinansowanie_inwestycyjne</th>\n",
+       "      <th>Dochody_dofinansowanie_razem</th>\n",
+       "      <th>Udzialy_w_podatkach_dochodowych_od_osob_fizycznych</th>\n",
+       "      <th>Udzialy_w_podatkach_dochodowych_od_osob_prywatnych</th>\n",
+       "      <th>Udzialy_w_podatkach_dochodowych_razem</th>\n",
+       "      <th>Wplywy_z_innych_lokalnych_oplat</th>\n",
+       "      <th>Wplywy_z_oplaty_eksploatacyjnej</th>\n",
+       "      <th>Wplywy_z_oplaty_skarbowej</th>\n",
+       "      <th>Wplywy_z_oplaty_targowej</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2004</td>\n",
+       "      <td>5344205.00</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>184307.00</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>519209.00</td>\n",
+       "      <td>13285456.00</td>\n",
+       "      <td>1065169.00</td>\n",
+       "      <td>14350625.00</td>\n",
+       "      <td>44200.00</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>1209998.00</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2005</td>\n",
+       "      <td>4560489.00</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>96462.00</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>9024183.00</td>\n",
+       "      <td>15985331.00</td>\n",
+       "      <td>1170863.00</td>\n",
+       "      <td>17156194.00</td>\n",
+       "      <td>42840.00</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>1282943.00</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2006</td>\n",
+       "      <td>8528727.69</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>231470.96</td>\n",
+       "      <td>8752288.98</td>\n",
+       "      <td>8864860.57</td>\n",
+       "      <td>18101668.00</td>\n",
+       "      <td>1048115.83</td>\n",
+       "      <td>19149783.83</td>\n",
+       "      <td>37365.00</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>1203990.73</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2007</td>\n",
+       "      <td>15042480.34</td>\n",
+       "      <td>9219682.12</td>\n",
+       "      <td>339654.15</td>\n",
+       "      <td>18153240.30</td>\n",
+       "      <td>18438743.21</td>\n",
+       "      <td>21785308.00</td>\n",
+       "      <td>1336702.02</td>\n",
+       "      <td>23122010.02</td>\n",
+       "      <td>78798.51</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>1228704.53</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2008</td>\n",
+       "      <td>22797881.07</td>\n",
+       "      <td>9546379.31</td>\n",
+       "      <td>787256.69</td>\n",
+       "      <td>5046691.69</td>\n",
+       "      <td>5182137.79</td>\n",
+       "      <td>23974587.00</td>\n",
+       "      <td>1532633.44</td>\n",
+       "      <td>25507220.44</td>\n",
+       "      <td>83882.94</td>\n",
+       "      <td>NaN</td>\n",
+       "      <td>1364245.93</td>\n",
+       "      <td>NaN</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>47078</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2018</td>\n",
+       "      <td>16419859.31</td>\n",
+       "      <td>4261374.83</td>\n",
+       "      <td>1996824.80</td>\n",
+       "      <td>25285.92</td>\n",
+       "      <td>237485.34</td>\n",
+       "      <td>52799183.00</td>\n",
+       "      <td>2690098.17</td>\n",
+       "      <td>55489281.17</td>\n",
+       "      <td>10458871.30</td>\n",
+       "      <td>4684.54</td>\n",
+       "      <td>434077.88</td>\n",
+       "      <td>608625.90</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>47079</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2019</td>\n",
+       "      <td>8844350.07</td>\n",
+       "      <td>4324758.68</td>\n",
+       "      <td>2187576.47</td>\n",
+       "      <td>0.00</td>\n",
+       "      <td>225831.84</td>\n",
+       "      <td>55319040.00</td>\n",
+       "      <td>2770684.17</td>\n",
+       "      <td>58089724.17</td>\n",
+       "      <td>11369287.11</td>\n",
+       "      <td>3456.95</td>\n",
+       "      <td>415686.53</td>\n",
+       "      <td>610059.50</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>47080</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2020</td>\n",
+       "      <td>13485033.97</td>\n",
+       "      <td>6159923.01</td>\n",
+       "      <td>1917372.55</td>\n",
+       "      <td>21002107.00</td>\n",
+       "      <td>21192313.05</td>\n",
+       "      <td>53739656.00</td>\n",
+       "      <td>3144444.38</td>\n",
+       "      <td>56884100.38</td>\n",
+       "      <td>12281916.71</td>\n",
+       "      <td>5157.50</td>\n",
+       "      <td>355201.29</td>\n",
+       "      <td>507341.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>47081</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2021</td>\n",
+       "      <td>16928500.75</td>\n",
+       "      <td>7582499.62</td>\n",
+       "      <td>4110105.72</td>\n",
+       "      <td>888293.63</td>\n",
+       "      <td>1072910.83</td>\n",
+       "      <td>63936763.00</td>\n",
+       "      <td>3975531.95</td>\n",
+       "      <td>67912294.95</td>\n",
+       "      <td>17127683.55</td>\n",
+       "      <td>27746.70</td>\n",
+       "      <td>416473.03</td>\n",
+       "      <td>0.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>47082</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2022</td>\n",
+       "      <td>30415536.99</td>\n",
+       "      <td>8651170.05</td>\n",
+       "      <td>4117086.30</td>\n",
+       "      <td>207597.50</td>\n",
+       "      <td>800347.63</td>\n",
+       "      <td>64657287.40</td>\n",
+       "      <td>4082611.64</td>\n",
+       "      <td>68739899.04</td>\n",
+       "      <td>19150342.25</td>\n",
+       "      <td>5035.87</td>\n",
+       "      <td>421424.91</td>\n",
+       "      <td>1233266.30</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>47083 rows × 14 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "Rodzaje dochodów      Kod   Rok  Dochody_z_majatku   \n",
+       "0                 0201011  2004         5344205.00  \\\n",
+       "1                 0201011  2005         4560489.00   \n",
+       "2                 0201011  2006         8528727.69   \n",
+       "3                 0201011  2007        15042480.34   \n",
+       "4                 0201011  2008        22797881.07   \n",
+       "...                   ...   ...                ...   \n",
+       "47078             3263011  2018        16419859.31   \n",
+       "47079             3263011  2019         8844350.07   \n",
+       "47080             3263011  2020        13485033.97   \n",
+       "47081             3263011  2021        16928500.75   \n",
+       "47082             3263011  2022        30415536.99   \n",
+       "\n",
+       "Rodzaje dochodów  Dochody_z_najmu_i_dzierzawy  Dochody_z_uslug   \n",
+       "0                                         NaN        184307.00  \\\n",
+       "1                                         NaN         96462.00   \n",
+       "2                                         NaN        231470.96   \n",
+       "3                                  9219682.12        339654.15   \n",
+       "4                                  9546379.31        787256.69   \n",
+       "...                                       ...              ...   \n",
+       "47078                              4261374.83       1996824.80   \n",
+       "47079                              4324758.68       2187576.47   \n",
+       "47080                              6159923.01       1917372.55   \n",
+       "47081                              7582499.62       4110105.72   \n",
+       "47082                              8651170.05       4117086.30   \n",
+       "\n",
+       "Rodzaje dochodów  Dochody_dofinansowanie_inwestycyjne   \n",
+       "0                                                 NaN  \\\n",
+       "1                                                 NaN   \n",
+       "2                                          8752288.98   \n",
+       "3                                         18153240.30   \n",
+       "4                                          5046691.69   \n",
+       "...                                               ...   \n",
+       "47078                                        25285.92   \n",
+       "47079                                            0.00   \n",
+       "47080                                     21002107.00   \n",
+       "47081                                       888293.63   \n",
+       "47082                                       207597.50   \n",
+       "\n",
+       "Rodzaje dochodów  Dochody_dofinansowanie_razem   \n",
+       "0                                    519209.00  \\\n",
+       "1                                   9024183.00   \n",
+       "2                                   8864860.57   \n",
+       "3                                  18438743.21   \n",
+       "4                                   5182137.79   \n",
+       "...                                        ...   \n",
+       "47078                                237485.34   \n",
+       "47079                                225831.84   \n",
+       "47080                              21192313.05   \n",
+       "47081                               1072910.83   \n",
+       "47082                                800347.63   \n",
+       "\n",
+       "Rodzaje dochodów  Udzialy_w_podatkach_dochodowych_od_osob_fizycznych   \n",
+       "0                                                       13285456.00   \\\n",
+       "1                                                       15985331.00    \n",
+       "2                                                       18101668.00    \n",
+       "3                                                       21785308.00    \n",
+       "4                                                       23974587.00    \n",
+       "...                                                             ...    \n",
+       "47078                                                   52799183.00    \n",
+       "47079                                                   55319040.00    \n",
+       "47080                                                   53739656.00    \n",
+       "47081                                                   63936763.00    \n",
+       "47082                                                   64657287.40    \n",
+       "\n",
+       "Rodzaje dochodów  Udzialy_w_podatkach_dochodowych_od_osob_prywatnych   \n",
+       "0                                                        1065169.00   \\\n",
+       "1                                                        1170863.00    \n",
+       "2                                                        1048115.83    \n",
+       "3                                                        1336702.02    \n",
+       "4                                                        1532633.44    \n",
+       "...                                                             ...    \n",
+       "47078                                                    2690098.17    \n",
+       "47079                                                    2770684.17    \n",
+       "47080                                                    3144444.38    \n",
+       "47081                                                    3975531.95    \n",
+       "47082                                                    4082611.64    \n",
+       "\n",
+       "Rodzaje dochodów  Udzialy_w_podatkach_dochodowych_razem   \n",
+       "0                                           14350625.00  \\\n",
+       "1                                           17156194.00   \n",
+       "2                                           19149783.83   \n",
+       "3                                           23122010.02   \n",
+       "4                                           25507220.44   \n",
+       "...                                                 ...   \n",
+       "47078                                       55489281.17   \n",
+       "47079                                       58089724.17   \n",
+       "47080                                       56884100.38   \n",
+       "47081                                       67912294.95   \n",
+       "47082                                       68739899.04   \n",
+       "\n",
+       "Rodzaje dochodów  Wplywy_z_innych_lokalnych_oplat   \n",
+       "0                                        44200.00  \\\n",
+       "1                                        42840.00   \n",
+       "2                                        37365.00   \n",
+       "3                                        78798.51   \n",
+       "4                                        83882.94   \n",
+       "...                                           ...   \n",
+       "47078                                 10458871.30   \n",
+       "47079                                 11369287.11   \n",
+       "47080                                 12281916.71   \n",
+       "47081                                 17127683.55   \n",
+       "47082                                 19150342.25   \n",
+       "\n",
+       "Rodzaje dochodów  Wplywy_z_oplaty_eksploatacyjnej  Wplywy_z_oplaty_skarbowej   \n",
+       "0                                             NaN                 1209998.00  \\\n",
+       "1                                             NaN                 1282943.00   \n",
+       "2                                             NaN                 1203990.73   \n",
+       "3                                             NaN                 1228704.53   \n",
+       "4                                             NaN                 1364245.93   \n",
+       "...                                           ...                        ...   \n",
+       "47078                                     4684.54                  434077.88   \n",
+       "47079                                     3456.95                  415686.53   \n",
+       "47080                                     5157.50                  355201.29   \n",
+       "47081                                    27746.70                  416473.03   \n",
+       "47082                                     5035.87                  421424.91   \n",
+       "\n",
+       "Rodzaje dochodów  Wplywy_z_oplaty_targowej  \n",
+       "0                                      NaN  \n",
+       "1                                      NaN  \n",
+       "2                                      NaN  \n",
+       "3                                      NaN  \n",
+       "4                                      NaN  \n",
+       "...                                    ...  \n",
+       "47078                            608625.90  \n",
+       "47079                            610059.50  \n",
+       "47080                            507341.00  \n",
+       "47081                                 0.00  \n",
+       "47082                           1233266.30  \n",
+       "\n",
+       "[47083 rows x 14 columns]"
+      ]
+     },
+     "execution_count": 28,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df_fina_2 = pd.read_csv(\n",
+    "    'FINA_2622_CREL_2.csv',\n",
+    "    sep=';',\n",
+    "    converters={'Kod': str},\n",
+    "    decimal=',')\n",
+    "df_fina_2 = df_fina_2[['Kod', 'Rodzaje dochodów', 'Rok', 'Wartosc']]\n",
+    "df_fina_2 = df_fina_2.dropna()\n",
+    "df_fina_2 = df_fina_2.pivot_table(index=['Kod', 'Rok'], columns='Rodzaje dochodów', values='Wartosc').reset_index()\n",
+    "df_fina_2 = df_fina_2.rename(columns={\n",
+    "    'dochody z majątku': 'Dochody_z_majatku',\n",
+    "    'dochody z majątku - dochody z najmu i dzierżawy składników majątkowych JST oraz innych umów o podobnym charakterze': 'Dochody_z_najmu_i_dzierzawy',\n",
+    "    'pozostałe dochody - wpływy z usług': 'Dochody_z_uslug',\n",
+    "    'pozostałe dochody - środki na dofinansowanie własnych zadań pozyskane z innych źródeł - inwestycyjne': 'Dochody_dofinansowanie_inwestycyjne',\n",
+    "    'pozostałe dochody - środki na dofinansowanie własnych zadań pozyskane z innych źródeł - razem': 'Dochody_dofinansowanie_razem',\n",
+    "    'udziały w podatkach stanowiących dochody budżetu państwa podatek dochodowy od osób fizycznych': 'Udzialy_w_podatkach_dochodowych_od_osob_fizycznych',\n",
+    "    'udziały w podatkach stanowiących dochody budżetu państwa podatek dochodowy od osób prawnych': 'Udzialy_w_podatkach_dochodowych_od_osob_prywatnych',\n",
+    "    'udziały w podatkach stanowiących dochody budżetu państwa razem': 'Udzialy_w_podatkach_dochodowych_razem',\n",
+    "    'wpływy z innych lokalnych opłat pobieranych przez jednostki samorządu terytorialnego na podstawie odrębnych ustaw': 'Wplywy_z_innych_lokalnych_oplat',\n",
+    "    'wpływy z opłaty eksploatacyjnej': 'Wplywy_z_oplaty_eksploatacyjnej',\n",
+    "    'wpływy z opłaty skarbowej': 'Wplywy_z_oplaty_skarbowej',\n",
+    "    'wpływy z opłaty targowej': 'Wplywy_z_oplaty_targowej'})\n",
+    "\n",
+    "df_fina_2"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "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>Wiek</th>\n",
+       "      <th>Kod</th>\n",
+       "      <th>Rok</th>\n",
+       "      <th>Ludnosc_ogolem</th>\n",
+       "      <th>Ludnosc_w_wieku_poprodukcyjnym</th>\n",
+       "      <th>Ludnosc_w_wieku_produkcyjnym</th>\n",
+       "      <th>Ludnosc_w_wieku_produkcyjnym_mobilnym</th>\n",
+       "      <th>Ludnosc_w_wieku_produkcyjnym_niemobilnym</th>\n",
+       "      <th>Ludnosc_w_wieku_przedprodukcyjnym</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2010</td>\n",
+       "      <td>40309.00</td>\n",
+       "      <td>7683.00</td>\n",
+       "      <td>26085.00</td>\n",
+       "      <td>15183.00</td>\n",
+       "      <td>10902.00</td>\n",
+       "      <td>6541.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2011</td>\n",
+       "      <td>40119.00</td>\n",
+       "      <td>8020.00</td>\n",
+       "      <td>25647.00</td>\n",
+       "      <td>15047.00</td>\n",
+       "      <td>10600.00</td>\n",
+       "      <td>6452.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2012</td>\n",
+       "      <td>39851.00</td>\n",
+       "      <td>8392.00</td>\n",
+       "      <td>25160.00</td>\n",
+       "      <td>14932.00</td>\n",
+       "      <td>10228.00</td>\n",
+       "      <td>6299.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2013</td>\n",
+       "      <td>39603.00</td>\n",
+       "      <td>8678.00</td>\n",
+       "      <td>24720.00</td>\n",
+       "      <td>14784.00</td>\n",
+       "      <td>9936.00</td>\n",
+       "      <td>6205.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2014</td>\n",
+       "      <td>39464.00</td>\n",
+       "      <td>8971.00</td>\n",
+       "      <td>24307.00</td>\n",
+       "      <td>14645.00</td>\n",
+       "      <td>9662.00</td>\n",
+       "      <td>6186.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>48606</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2018</td>\n",
+       "      <td>40910.00</td>\n",
+       "      <td>10472.00</td>\n",
+       "      <td>24549.00</td>\n",
+       "      <td>14683.00</td>\n",
+       "      <td>9866.00</td>\n",
+       "      <td>5889.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>48607</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2019</td>\n",
+       "      <td>40888.00</td>\n",
+       "      <td>10788.00</td>\n",
+       "      <td>24209.00</td>\n",
+       "      <td>14429.00</td>\n",
+       "      <td>9780.00</td>\n",
+       "      <td>5891.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>48608</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2020</td>\n",
+       "      <td>40326.00</td>\n",
+       "      <td>10962.00</td>\n",
+       "      <td>23544.00</td>\n",
+       "      <td>13798.00</td>\n",
+       "      <td>9746.00</td>\n",
+       "      <td>5820.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>48609</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2021</td>\n",
+       "      <td>39834.00</td>\n",
+       "      <td>11050.00</td>\n",
+       "      <td>22976.00</td>\n",
+       "      <td>13277.00</td>\n",
+       "      <td>9699.00</td>\n",
+       "      <td>5808.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>48610</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2022</td>\n",
+       "      <td>39368.00</td>\n",
+       "      <td>11157.00</td>\n",
+       "      <td>22486.00</td>\n",
+       "      <td>12802.00</td>\n",
+       "      <td>9684.00</td>\n",
+       "      <td>5725.00</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>48611 rows × 8 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "Wiek       Kod   Rok  Ludnosc_ogolem  Ludnosc_w_wieku_poprodukcyjnym   \n",
+       "0      0201011  2010        40309.00                         7683.00  \\\n",
+       "1      0201011  2011        40119.00                         8020.00   \n",
+       "2      0201011  2012        39851.00                         8392.00   \n",
+       "3      0201011  2013        39603.00                         8678.00   \n",
+       "4      0201011  2014        39464.00                         8971.00   \n",
+       "...        ...   ...             ...                             ...   \n",
+       "48606  3263011  2018        40910.00                        10472.00   \n",
+       "48607  3263011  2019        40888.00                        10788.00   \n",
+       "48608  3263011  2020        40326.00                        10962.00   \n",
+       "48609  3263011  2021        39834.00                        11050.00   \n",
+       "48610  3263011  2022        39368.00                        11157.00   \n",
+       "\n",
+       "Wiek   Ludnosc_w_wieku_produkcyjnym  Ludnosc_w_wieku_produkcyjnym_mobilnym   \n",
+       "0                          26085.00                               15183.00  \\\n",
+       "1                          25647.00                               15047.00   \n",
+       "2                          25160.00                               14932.00   \n",
+       "3                          24720.00                               14784.00   \n",
+       "4                          24307.00                               14645.00   \n",
+       "...                             ...                                    ...   \n",
+       "48606                      24549.00                               14683.00   \n",
+       "48607                      24209.00                               14429.00   \n",
+       "48608                      23544.00                               13798.00   \n",
+       "48609                      22976.00                               13277.00   \n",
+       "48610                      22486.00                               12802.00   \n",
+       "\n",
+       "Wiek   Ludnosc_w_wieku_produkcyjnym_niemobilnym   \n",
+       "0                                      10902.00  \\\n",
+       "1                                      10600.00   \n",
+       "2                                      10228.00   \n",
+       "3                                       9936.00   \n",
+       "4                                       9662.00   \n",
+       "...                                         ...   \n",
+       "48606                                   9866.00   \n",
+       "48607                                   9780.00   \n",
+       "48608                                   9746.00   \n",
+       "48609                                   9699.00   \n",
+       "48610                                   9684.00   \n",
+       "\n",
+       "Wiek   Ludnosc_w_wieku_przedprodukcyjnym  \n",
+       "0                                6541.00  \n",
+       "1                                6452.00  \n",
+       "2                                6299.00  \n",
+       "3                                6205.00  \n",
+       "4                                6186.00  \n",
+       "...                                  ...  \n",
+       "48606                            5889.00  \n",
+       "48607                            5891.00  \n",
+       "48608                            5820.00  \n",
+       "48609                            5808.00  \n",
+       "48610                            5725.00  \n",
+       "\n",
+       "[48611 rows x 8 columns]"
+      ]
+     },
+     "execution_count": 29,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df_ludn_1 = pd.read_csv(  # ogolem\n",
+    "    'LUDN_1342_CREL_1.csv',\n",
+    "    sep=';',\n",
+    "    converters={'Kod': str},\n",
+    "    decimal=',')\n",
+    "df_ludn_1 = df_ludn_1[['Kod', 'Wiek', 'Rok', 'Wartosc']]\n",
+    "df_ludn_1 = df_ludn_1.dropna()\n",
+    "df_ludn_1 = df_ludn_1.pivot_table(index=['Kod', 'Rok'], columns='Wiek', values='Wartosc').reset_index()\n",
+    "df_ludn_1 = df_ludn_1.rename(columns={\n",
+    "    'ogółem': 'Ludnosc_ogolem',\n",
+    "    'w wieku poprodukcyjnym': 'Ludnosc_w_wieku_poprodukcyjnym',\n",
+    "    'w wieku produkcyjnym': 'Ludnosc_w_wieku_produkcyjnym',\n",
+    "    'w wieku produkcyjnym mobilnym': 'Ludnosc_w_wieku_produkcyjnym_mobilnym',\n",
+    "    'w wieku produkcyjnym niemobilnym': 'Ludnosc_w_wieku_produkcyjnym_niemobilnym',\n",
+    "    'w wieku przedprodukcyjnym': 'Ludnosc_w_wieku_przedprodukcyjnym'})\n",
+    "\n",
+    "df_ludn_1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 30,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "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>Wiek</th>\n",
+       "      <th>Kod</th>\n",
+       "      <th>Rok</th>\n",
+       "      <th>Ludnosc_mezczyzni</th>\n",
+       "      <th>Ludnosc_mezczyzni_w_wieku_poprodukcyjnym</th>\n",
+       "      <th>Ludnosc_mezczyzni_w_wieku_produkcyjnym</th>\n",
+       "      <th>Ludnosc_mezczyzni_w_wieku_produkcyjnym_mobilnym</th>\n",
+       "      <th>Ludnosc_mezczyzni_w_wieku_produkcyjnym_niemobilnym</th>\n",
+       "      <th>Ludnosc_mezczyzni_w_wieku_przedprodukcyjnym</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2010</td>\n",
+       "      <td>19085.00</td>\n",
+       "      <td>2153.00</td>\n",
+       "      <td>13535.00</td>\n",
+       "      <td>7720.00</td>\n",
+       "      <td>5815.00</td>\n",
+       "      <td>3397.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2011</td>\n",
+       "      <td>18985.00</td>\n",
+       "      <td>2222.00</td>\n",
+       "      <td>13398.00</td>\n",
+       "      <td>7647.00</td>\n",
+       "      <td>5751.00</td>\n",
+       "      <td>3365.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2012</td>\n",
+       "      <td>18859.00</td>\n",
+       "      <td>2370.00</td>\n",
+       "      <td>13238.00</td>\n",
+       "      <td>7611.00</td>\n",
+       "      <td>5627.00</td>\n",
+       "      <td>3251.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2013</td>\n",
+       "      <td>18737.00</td>\n",
+       "      <td>2477.00</td>\n",
+       "      <td>13028.00</td>\n",
+       "      <td>7501.00</td>\n",
+       "      <td>5527.00</td>\n",
+       "      <td>3232.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2014</td>\n",
+       "      <td>18640.00</td>\n",
+       "      <td>2620.00</td>\n",
+       "      <td>12832.00</td>\n",
+       "      <td>7442.00</td>\n",
+       "      <td>5390.00</td>\n",
+       "      <td>3188.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>48606</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2018</td>\n",
+       "      <td>19690.00</td>\n",
+       "      <td>3501.00</td>\n",
+       "      <td>13202.00</td>\n",
+       "      <td>7547.00</td>\n",
+       "      <td>5655.00</td>\n",
+       "      <td>2987.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>48607</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2019</td>\n",
+       "      <td>19683.00</td>\n",
+       "      <td>3644.00</td>\n",
+       "      <td>13044.00</td>\n",
+       "      <td>7417.00</td>\n",
+       "      <td>5627.00</td>\n",
+       "      <td>2995.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>48608</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2020</td>\n",
+       "      <td>19356.00</td>\n",
+       "      <td>3749.00</td>\n",
+       "      <td>12617.00</td>\n",
+       "      <td>6986.00</td>\n",
+       "      <td>5631.00</td>\n",
+       "      <td>2990.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>48609</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2021</td>\n",
+       "      <td>19096.00</td>\n",
+       "      <td>3852.00</td>\n",
+       "      <td>12267.00</td>\n",
+       "      <td>6747.00</td>\n",
+       "      <td>5520.00</td>\n",
+       "      <td>2977.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>48610</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2022</td>\n",
+       "      <td>18869.00</td>\n",
+       "      <td>3901.00</td>\n",
+       "      <td>12009.00</td>\n",
+       "      <td>6485.00</td>\n",
+       "      <td>5524.00</td>\n",
+       "      <td>2959.00</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>48611 rows × 8 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "Wiek       Kod   Rok  Ludnosc_mezczyzni   \n",
+       "0      0201011  2010           19085.00  \\\n",
+       "1      0201011  2011           18985.00   \n",
+       "2      0201011  2012           18859.00   \n",
+       "3      0201011  2013           18737.00   \n",
+       "4      0201011  2014           18640.00   \n",
+       "...        ...   ...                ...   \n",
+       "48606  3263011  2018           19690.00   \n",
+       "48607  3263011  2019           19683.00   \n",
+       "48608  3263011  2020           19356.00   \n",
+       "48609  3263011  2021           19096.00   \n",
+       "48610  3263011  2022           18869.00   \n",
+       "\n",
+       "Wiek   Ludnosc_mezczyzni_w_wieku_poprodukcyjnym   \n",
+       "0                                       2153.00  \\\n",
+       "1                                       2222.00   \n",
+       "2                                       2370.00   \n",
+       "3                                       2477.00   \n",
+       "4                                       2620.00   \n",
+       "...                                         ...   \n",
+       "48606                                   3501.00   \n",
+       "48607                                   3644.00   \n",
+       "48608                                   3749.00   \n",
+       "48609                                   3852.00   \n",
+       "48610                                   3901.00   \n",
+       "\n",
+       "Wiek   Ludnosc_mezczyzni_w_wieku_produkcyjnym   \n",
+       "0                                    13535.00  \\\n",
+       "1                                    13398.00   \n",
+       "2                                    13238.00   \n",
+       "3                                    13028.00   \n",
+       "4                                    12832.00   \n",
+       "...                                       ...   \n",
+       "48606                                13202.00   \n",
+       "48607                                13044.00   \n",
+       "48608                                12617.00   \n",
+       "48609                                12267.00   \n",
+       "48610                                12009.00   \n",
+       "\n",
+       "Wiek   Ludnosc_mezczyzni_w_wieku_produkcyjnym_mobilnym   \n",
+       "0                                              7720.00  \\\n",
+       "1                                              7647.00   \n",
+       "2                                              7611.00   \n",
+       "3                                              7501.00   \n",
+       "4                                              7442.00   \n",
+       "...                                                ...   \n",
+       "48606                                          7547.00   \n",
+       "48607                                          7417.00   \n",
+       "48608                                          6986.00   \n",
+       "48609                                          6747.00   \n",
+       "48610                                          6485.00   \n",
+       "\n",
+       "Wiek   Ludnosc_mezczyzni_w_wieku_produkcyjnym_niemobilnym   \n",
+       "0                                                5815.00   \\\n",
+       "1                                                5751.00    \n",
+       "2                                                5627.00    \n",
+       "3                                                5527.00    \n",
+       "4                                                5390.00    \n",
+       "...                                                  ...    \n",
+       "48606                                            5655.00    \n",
+       "48607                                            5627.00    \n",
+       "48608                                            5631.00    \n",
+       "48609                                            5520.00    \n",
+       "48610                                            5524.00    \n",
+       "\n",
+       "Wiek   Ludnosc_mezczyzni_w_wieku_przedprodukcyjnym  \n",
+       "0                                          3397.00  \n",
+       "1                                          3365.00  \n",
+       "2                                          3251.00  \n",
+       "3                                          3232.00  \n",
+       "4                                          3188.00  \n",
+       "...                                            ...  \n",
+       "48606                                      2987.00  \n",
+       "48607                                      2995.00  \n",
+       "48608                                      2990.00  \n",
+       "48609                                      2977.00  \n",
+       "48610                                      2959.00  \n",
+       "\n",
+       "[48611 rows x 8 columns]"
+      ]
+     },
+     "execution_count": 30,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df_ludn_2 = pd.read_csv(  # mezczyzni\n",
+    "    'LUDN_1342_CREL_2.csv',\n",
+    "    sep=';',\n",
+    "    converters={'Kod': str},\n",
+    "    decimal=',')\n",
+    "df_ludn_2 = df_ludn_2[['Kod', 'Wiek', 'Rok', 'Wartosc']]\n",
+    "df_ludn_2 = df_ludn_2.dropna()\n",
+    "df_ludn_2 = df_ludn_2.pivot_table(index=['Kod', 'Rok'], columns='Wiek', values='Wartosc').reset_index()\n",
+    "df_ludn_2 = df_ludn_2.rename(columns={\n",
+    "    'ogółem': 'Ludnosc_mezczyzni',\n",
+    "    'w wieku poprodukcyjnym': 'Ludnosc_mezczyzni_w_wieku_poprodukcyjnym',\n",
+    "    'w wieku produkcyjnym': 'Ludnosc_mezczyzni_w_wieku_produkcyjnym',\n",
+    "    'w wieku produkcyjnym mobilnym': 'Ludnosc_mezczyzni_w_wieku_produkcyjnym_mobilnym',\n",
+    "    'w wieku produkcyjnym niemobilnym': 'Ludnosc_mezczyzni_w_wieku_produkcyjnym_niemobilnym',\n",
+    "    'w wieku przedprodukcyjnym': 'Ludnosc_mezczyzni_w_wieku_przedprodukcyjnym'})\n",
+    "\n",
+    "df_ludn_2"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 31,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "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>Wiek</th>\n",
+       "      <th>Kod</th>\n",
+       "      <th>Rok</th>\n",
+       "      <th>Ludnosc_kobiety</th>\n",
+       "      <th>Ludnosc_kobiety_w_wieku_poprodukcyjnym</th>\n",
+       "      <th>Ludnosc_kobiety_w_wieku_produkcyjnym</th>\n",
+       "      <th>Ludnosc_kobiety_w_wieku_produkcyjnym_mobilnym</th>\n",
+       "      <th>Ludnosc_kobiety_w_wieku_produkcyjnym_niemobilnym</th>\n",
+       "      <th>Ludnosc_kobiety_w_wieku_przedprodukcyjnym</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2010</td>\n",
+       "      <td>21224.00</td>\n",
+       "      <td>5530.00</td>\n",
+       "      <td>12550.00</td>\n",
+       "      <td>7463.00</td>\n",
+       "      <td>5087.00</td>\n",
+       "      <td>3144.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2011</td>\n",
+       "      <td>21134.00</td>\n",
+       "      <td>5798.00</td>\n",
+       "      <td>12249.00</td>\n",
+       "      <td>7400.00</td>\n",
+       "      <td>4849.00</td>\n",
+       "      <td>3087.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2012</td>\n",
+       "      <td>20992.00</td>\n",
+       "      <td>6022.00</td>\n",
+       "      <td>11922.00</td>\n",
+       "      <td>7321.00</td>\n",
+       "      <td>4601.00</td>\n",
+       "      <td>3048.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2013</td>\n",
+       "      <td>20866.00</td>\n",
+       "      <td>6201.00</td>\n",
+       "      <td>11692.00</td>\n",
+       "      <td>7283.00</td>\n",
+       "      <td>4409.00</td>\n",
+       "      <td>2973.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2014</td>\n",
+       "      <td>20824.00</td>\n",
+       "      <td>6351.00</td>\n",
+       "      <td>11475.00</td>\n",
+       "      <td>7203.00</td>\n",
+       "      <td>4272.00</td>\n",
+       "      <td>2998.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>48606</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2018</td>\n",
+       "      <td>21220.00</td>\n",
+       "      <td>6971.00</td>\n",
+       "      <td>11347.00</td>\n",
+       "      <td>7136.00</td>\n",
+       "      <td>4211.00</td>\n",
+       "      <td>2902.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>48607</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2019</td>\n",
+       "      <td>21205.00</td>\n",
+       "      <td>7144.00</td>\n",
+       "      <td>11165.00</td>\n",
+       "      <td>7012.00</td>\n",
+       "      <td>4153.00</td>\n",
+       "      <td>2896.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>48608</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2020</td>\n",
+       "      <td>20970.00</td>\n",
+       "      <td>7213.00</td>\n",
+       "      <td>10927.00</td>\n",
+       "      <td>6812.00</td>\n",
+       "      <td>4115.00</td>\n",
+       "      <td>2830.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>48609</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2021</td>\n",
+       "      <td>20738.00</td>\n",
+       "      <td>7198.00</td>\n",
+       "      <td>10709.00</td>\n",
+       "      <td>6530.00</td>\n",
+       "      <td>4179.00</td>\n",
+       "      <td>2831.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>48610</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2022</td>\n",
+       "      <td>20499.00</td>\n",
+       "      <td>7256.00</td>\n",
+       "      <td>10477.00</td>\n",
+       "      <td>6317.00</td>\n",
+       "      <td>4160.00</td>\n",
+       "      <td>2766.00</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>48611 rows × 8 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "Wiek       Kod   Rok  Ludnosc_kobiety  Ludnosc_kobiety_w_wieku_poprodukcyjnym   \n",
+       "0      0201011  2010         21224.00                                 5530.00  \\\n",
+       "1      0201011  2011         21134.00                                 5798.00   \n",
+       "2      0201011  2012         20992.00                                 6022.00   \n",
+       "3      0201011  2013         20866.00                                 6201.00   \n",
+       "4      0201011  2014         20824.00                                 6351.00   \n",
+       "...        ...   ...              ...                                     ...   \n",
+       "48606  3263011  2018         21220.00                                 6971.00   \n",
+       "48607  3263011  2019         21205.00                                 7144.00   \n",
+       "48608  3263011  2020         20970.00                                 7213.00   \n",
+       "48609  3263011  2021         20738.00                                 7198.00   \n",
+       "48610  3263011  2022         20499.00                                 7256.00   \n",
+       "\n",
+       "Wiek   Ludnosc_kobiety_w_wieku_produkcyjnym   \n",
+       "0                                  12550.00  \\\n",
+       "1                                  12249.00   \n",
+       "2                                  11922.00   \n",
+       "3                                  11692.00   \n",
+       "4                                  11475.00   \n",
+       "...                                     ...   \n",
+       "48606                              11347.00   \n",
+       "48607                              11165.00   \n",
+       "48608                              10927.00   \n",
+       "48609                              10709.00   \n",
+       "48610                              10477.00   \n",
+       "\n",
+       "Wiek   Ludnosc_kobiety_w_wieku_produkcyjnym_mobilnym   \n",
+       "0                                            7463.00  \\\n",
+       "1                                            7400.00   \n",
+       "2                                            7321.00   \n",
+       "3                                            7283.00   \n",
+       "4                                            7203.00   \n",
+       "...                                              ...   \n",
+       "48606                                        7136.00   \n",
+       "48607                                        7012.00   \n",
+       "48608                                        6812.00   \n",
+       "48609                                        6530.00   \n",
+       "48610                                        6317.00   \n",
+       "\n",
+       "Wiek   Ludnosc_kobiety_w_wieku_produkcyjnym_niemobilnym   \n",
+       "0                                               5087.00  \\\n",
+       "1                                               4849.00   \n",
+       "2                                               4601.00   \n",
+       "3                                               4409.00   \n",
+       "4                                               4272.00   \n",
+       "...                                                 ...   \n",
+       "48606                                           4211.00   \n",
+       "48607                                           4153.00   \n",
+       "48608                                           4115.00   \n",
+       "48609                                           4179.00   \n",
+       "48610                                           4160.00   \n",
+       "\n",
+       "Wiek   Ludnosc_kobiety_w_wieku_przedprodukcyjnym  \n",
+       "0                                        3144.00  \n",
+       "1                                        3087.00  \n",
+       "2                                        3048.00  \n",
+       "3                                        2973.00  \n",
+       "4                                        2998.00  \n",
+       "...                                          ...  \n",
+       "48606                                    2902.00  \n",
+       "48607                                    2896.00  \n",
+       "48608                                    2830.00  \n",
+       "48609                                    2831.00  \n",
+       "48610                                    2766.00  \n",
+       "\n",
+       "[48611 rows x 8 columns]"
+      ]
+     },
+     "execution_count": 31,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df_ludn_3 = pd.read_csv(  # kobiety\n",
+    "    'LUDN_1342_CREL_3.csv',\n",
+    "    sep=';',\n",
+    "    converters={'Kod': str},\n",
+    "    decimal=',')\n",
+    "df_ludn_3 = df_ludn_3[['Kod', 'Wiek', 'Rok', 'Wartosc']]\n",
+    "df_ludn_3 = df_ludn_3.dropna()\n",
+    "df_ludn_3 = df_ludn_3.pivot_table(index=['Kod', 'Rok'], columns='Wiek', values='Wartosc').reset_index()\n",
+    "df_ludn_3 = df_ludn_3.rename(columns={\n",
+    "    'ogółem': 'Ludnosc_kobiety',\n",
+    "    'w wieku poprodukcyjnym': 'Ludnosc_kobiety_w_wieku_poprodukcyjnym',\n",
+    "    'w wieku produkcyjnym': 'Ludnosc_kobiety_w_wieku_produkcyjnym',\n",
+    "    'w wieku produkcyjnym mobilnym': 'Ludnosc_kobiety_w_wieku_produkcyjnym_mobilnym',\n",
+    "    'w wieku produkcyjnym niemobilnym': 'Ludnosc_kobiety_w_wieku_produkcyjnym_niemobilnym',\n",
+    "    'w wieku przedprodukcyjnym': 'Ludnosc_kobiety_w_wieku_przedprodukcyjnym'})\n",
+    "\n",
+    "df_ludn_3"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "..."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 32,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df_data = df_dofinansowanie_agg.copy()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 33,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df_data = df_data.merge(df_podz, left_on=[df_data['Kod'].str.slice(stop=-1), 'Rok'], right_on=[df_podz['Kod'].str.slice(stop=-1), 'Rok'], how='left', suffixes=(None, '_podz'))\n",
+    "df_data = df_data.drop(['key_0', 'Kod_podz'], axis=1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 34,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df_data = df_data.merge(df_wyna, left_on=[df_data['Kod'].str.slice(stop=-3), 'Rok'], right_on=[df_wyna['Kod'].str.slice(stop=-3), 'Rok'], how='left', suffixes=(None, '_wyna'))\n",
+    "df_data = df_data.drop(['key_0', 'Kod_wyna'], axis=1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 35,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df_data = df_data.merge(df_fina_1, left_on=[df_data['Kod'].str.slice(stop=-1), 'Rok'], right_on=[df_fina_1['Kod'].str.slice(stop=-1), 'Rok'], how='left', suffixes=(None, '_fina_1'))\n",
+    "df_data = df_data.drop(['key_0', 'Kod_fina_1'], axis=1)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 36,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df_data = df_data.merge(df_fina_2, left_on=[df_data['Kod'].str.slice(stop=-1), 'Rok'], right_on=[df_fina_2['Kod'].str.slice(stop=-1), 'Rok'], how='left', suffixes=(None, '_fina_2'))\n",
+    "df_data = df_data.drop(['key_0', 'Kod_fina_2'], axis=1)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "..."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 37,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "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>Kod</th>\n",
+       "      <th>Rok</th>\n",
+       "      <th>Suma</th>\n",
+       "      <th>Powierzchnia</th>\n",
+       "      <th>Wynagrodzenie_ogolem</th>\n",
+       "      <th>Wynagrodzenie_w_relacji_do_sredniej</th>\n",
+       "      <th>Dochody_podatek_lesny</th>\n",
+       "      <th>Dochody_podatek_PCC</th>\n",
+       "      <th>Dochody_podatek_...</th>\n",
+       "      <th>Dochody_podatek_od_nieruchomosci</th>\n",
+       "      <th>...</th>\n",
+       "      <th>Dochody_z_uslug</th>\n",
+       "      <th>Dochody_dofinansowanie_inwestycyjne</th>\n",
+       "      <th>Dochody_dofinansowanie_razem</th>\n",
+       "      <th>Udzialy_w_podatkach_dochodowych_od_osob_fizycznych</th>\n",
+       "      <th>Udzialy_w_podatkach_dochodowych_od_osob_prywatnych</th>\n",
+       "      <th>Udzialy_w_podatkach_dochodowych_razem</th>\n",
+       "      <th>Wplywy_z_innych_lokalnych_oplat</th>\n",
+       "      <th>Wplywy_z_oplaty_eksploatacyjnej</th>\n",
+       "      <th>Wplywy_z_oplaty_skarbowej</th>\n",
+       "      <th>Wplywy_z_oplaty_targowej</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2014</td>\n",
+       "      <td>2482706.36</td>\n",
+       "      <td>24.00</td>\n",
+       "      <td>3422.90</td>\n",
+       "      <td>85.50</td>\n",
+       "      <td>8755.97</td>\n",
+       "      <td>1530935.80</td>\n",
+       "      <td>84220.54</td>\n",
+       "      <td>23823448.76</td>\n",
+       "      <td>...</td>\n",
+       "      <td>2380938.91</td>\n",
+       "      <td>61031.32</td>\n",
+       "      <td>1900308.99</td>\n",
+       "      <td>27297399.00</td>\n",
+       "      <td>1984447.80</td>\n",
+       "      <td>29281846.80</td>\n",
+       "      <td>7018623.68</td>\n",
+       "      <td>28867.32</td>\n",
+       "      <td>928926.81</td>\n",
+       "      <td>188736.80</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2016</td>\n",
+       "      <td>1016449.03</td>\n",
+       "      <td>24.00</td>\n",
+       "      <td>3666.19</td>\n",
+       "      <td>85.40</td>\n",
+       "      <td>14900.55</td>\n",
+       "      <td>1631451.83</td>\n",
+       "      <td>77726.70</td>\n",
+       "      <td>21961009.53</td>\n",
+       "      <td>...</td>\n",
+       "      <td>3748786.27</td>\n",
+       "      <td>111299.61</td>\n",
+       "      <td>284981.40</td>\n",
+       "      <td>31878605.00</td>\n",
+       "      <td>3726450.74</td>\n",
+       "      <td>35605055.74</td>\n",
+       "      <td>7277990.62</td>\n",
+       "      <td>31466.01</td>\n",
+       "      <td>475771.76</td>\n",
+       "      <td>158803.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2017</td>\n",
+       "      <td>1507540.44</td>\n",
+       "      <td>24.00</td>\n",
+       "      <td>3901.94</td>\n",
+       "      <td>86.20</td>\n",
+       "      <td>14988.07</td>\n",
+       "      <td>2022884.74</td>\n",
+       "      <td>66169.12</td>\n",
+       "      <td>22107840.97</td>\n",
+       "      <td>...</td>\n",
+       "      <td>4137582.84</td>\n",
+       "      <td>89949.86</td>\n",
+       "      <td>118699.74</td>\n",
+       "      <td>34702412.00</td>\n",
+       "      <td>2536379.13</td>\n",
+       "      <td>37238791.13</td>\n",
+       "      <td>7339238.14</td>\n",
+       "      <td>24998.04</td>\n",
+       "      <td>511127.33</td>\n",
+       "      <td>131562.90</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2018</td>\n",
+       "      <td>1125087.75</td>\n",
+       "      <td>24.00</td>\n",
+       "      <td>4214.66</td>\n",
+       "      <td>87.20</td>\n",
+       "      <td>15221.03</td>\n",
+       "      <td>2519471.86</td>\n",
+       "      <td>74351.81</td>\n",
+       "      <td>22767685.01</td>\n",
+       "      <td>...</td>\n",
+       "      <td>5556645.41</td>\n",
+       "      <td>0.00</td>\n",
+       "      <td>24676.92</td>\n",
+       "      <td>39128194.00</td>\n",
+       "      <td>1506913.18</td>\n",
+       "      <td>40635107.18</td>\n",
+       "      <td>7494838.88</td>\n",
+       "      <td>16194.57</td>\n",
+       "      <td>479489.69</td>\n",
+       "      <td>124935.70</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>0201011</td>\n",
+       "      <td>2020</td>\n",
+       "      <td>9052911.45</td>\n",
+       "      <td>24.00</td>\n",
+       "      <td>4887.95</td>\n",
+       "      <td>88.50</td>\n",
+       "      <td>15459.78</td>\n",
+       "      <td>2632509.99</td>\n",
+       "      <td>74425.14</td>\n",
+       "      <td>28336580.30</td>\n",
+       "      <td>...</td>\n",
+       "      <td>5344170.62</td>\n",
+       "      <td>2640028.00</td>\n",
+       "      <td>2660441.20</td>\n",
+       "      <td>40456699.00</td>\n",
+       "      <td>2143819.12</td>\n",
+       "      <td>42600518.12</td>\n",
+       "      <td>7627128.14</td>\n",
+       "      <td>5661.00</td>\n",
+       "      <td>440259.90</td>\n",
+       "      <td>62502.50</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9723</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2017</td>\n",
+       "      <td>467956789.99</td>\n",
+       "      <td>197.00</td>\n",
+       "      <td>4536.08</td>\n",
+       "      <td>100.20</td>\n",
+       "      <td>150527.43</td>\n",
+       "      <td>5163900.56</td>\n",
+       "      <td>587305.94</td>\n",
+       "      <td>74099232.11</td>\n",
+       "      <td>...</td>\n",
+       "      <td>1859435.58</td>\n",
+       "      <td>20000.18</td>\n",
+       "      <td>212570.70</td>\n",
+       "      <td>47760120.00</td>\n",
+       "      <td>1199589.08</td>\n",
+       "      <td>48959709.08</td>\n",
+       "      <td>10930683.15</td>\n",
+       "      <td>38789.78</td>\n",
+       "      <td>404331.07</td>\n",
+       "      <td>622509.60</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9724</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2018</td>\n",
+       "      <td>11972196.58</td>\n",
+       "      <td>202.00</td>\n",
+       "      <td>4779.09</td>\n",
+       "      <td>98.80</td>\n",
+       "      <td>154462.39</td>\n",
+       "      <td>5361951.37</td>\n",
+       "      <td>572868.36</td>\n",
+       "      <td>108107448.79</td>\n",
+       "      <td>...</td>\n",
+       "      <td>1996824.80</td>\n",
+       "      <td>25285.92</td>\n",
+       "      <td>237485.34</td>\n",
+       "      <td>52799183.00</td>\n",
+       "      <td>2690098.17</td>\n",
+       "      <td>55489281.17</td>\n",
+       "      <td>10458871.30</td>\n",
+       "      <td>4684.54</td>\n",
+       "      <td>434077.88</td>\n",
+       "      <td>608625.90</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9725</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2019</td>\n",
+       "      <td>2074346.69</td>\n",
+       "      <td>202.00</td>\n",
+       "      <td>5222.62</td>\n",
+       "      <td>100.80</td>\n",
+       "      <td>150329.31</td>\n",
+       "      <td>6088184.20</td>\n",
+       "      <td>468411.51</td>\n",
+       "      <td>38527846.59</td>\n",
+       "      <td>...</td>\n",
+       "      <td>2187576.47</td>\n",
+       "      <td>0.00</td>\n",
+       "      <td>225831.84</td>\n",
+       "      <td>55319040.00</td>\n",
+       "      <td>2770684.17</td>\n",
+       "      <td>58089724.17</td>\n",
+       "      <td>11369287.11</td>\n",
+       "      <td>3456.95</td>\n",
+       "      <td>415686.53</td>\n",
+       "      <td>610059.50</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9726</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2020</td>\n",
+       "      <td>8314185.11</td>\n",
+       "      <td>202.00</td>\n",
+       "      <td>5517.68</td>\n",
+       "      <td>99.90</td>\n",
+       "      <td>156556.52</td>\n",
+       "      <td>5125090.74</td>\n",
+       "      <td>329522.12</td>\n",
+       "      <td>78767466.83</td>\n",
+       "      <td>...</td>\n",
+       "      <td>1917372.55</td>\n",
+       "      <td>21002107.00</td>\n",
+       "      <td>21192313.05</td>\n",
+       "      <td>53739656.00</td>\n",
+       "      <td>3144444.38</td>\n",
+       "      <td>56884100.38</td>\n",
+       "      <td>12281916.71</td>\n",
+       "      <td>5157.50</td>\n",
+       "      <td>355201.29</td>\n",
+       "      <td>507341.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9727</th>\n",
+       "      <td>3263011</td>\n",
+       "      <td>2021</td>\n",
+       "      <td>752819.71</td>\n",
+       "      <td>202.00</td>\n",
+       "      <td>5564.41</td>\n",
+       "      <td>92.70</td>\n",
+       "      <td>163778.36</td>\n",
+       "      <td>9082482.28</td>\n",
+       "      <td>492045.28</td>\n",
+       "      <td>78491368.35</td>\n",
+       "      <td>...</td>\n",
+       "      <td>4110105.72</td>\n",
+       "      <td>888293.63</td>\n",
+       "      <td>1072910.83</td>\n",
+       "      <td>63936763.00</td>\n",
+       "      <td>3975531.95</td>\n",
+       "      <td>67912294.95</td>\n",
+       "      <td>17127683.55</td>\n",
+       "      <td>27746.70</td>\n",
+       "      <td>416473.03</td>\n",
+       "      <td>0.00</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>9728 rows × 27 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "          Kod   Rok         Suma  Powierzchnia  Wynagrodzenie_ogolem   \n",
+       "0     0201011  2014   2482706.36         24.00               3422.90  \\\n",
+       "1     0201011  2016   1016449.03         24.00               3666.19   \n",
+       "2     0201011  2017   1507540.44         24.00               3901.94   \n",
+       "3     0201011  2018   1125087.75         24.00               4214.66   \n",
+       "4     0201011  2020   9052911.45         24.00               4887.95   \n",
+       "...       ...   ...          ...           ...                   ...   \n",
+       "9723  3263011  2017 467956789.99        197.00               4536.08   \n",
+       "9724  3263011  2018  11972196.58        202.00               4779.09   \n",
+       "9725  3263011  2019   2074346.69        202.00               5222.62   \n",
+       "9726  3263011  2020   8314185.11        202.00               5517.68   \n",
+       "9727  3263011  2021    752819.71        202.00               5564.41   \n",
+       "\n",
+       "      Wynagrodzenie_w_relacji_do_sredniej  Dochody_podatek_lesny   \n",
+       "0                                   85.50                8755.97  \\\n",
+       "1                                   85.40               14900.55   \n",
+       "2                                   86.20               14988.07   \n",
+       "3                                   87.20               15221.03   \n",
+       "4                                   88.50               15459.78   \n",
+       "...                                   ...                    ...   \n",
+       "9723                               100.20              150527.43   \n",
+       "9724                                98.80              154462.39   \n",
+       "9725                               100.80              150329.31   \n",
+       "9726                                99.90              156556.52   \n",
+       "9727                                92.70              163778.36   \n",
+       "\n",
+       "      Dochody_podatek_PCC  Dochody_podatek_...   \n",
+       "0              1530935.80             84220.54  \\\n",
+       "1              1631451.83             77726.70   \n",
+       "2              2022884.74             66169.12   \n",
+       "3              2519471.86             74351.81   \n",
+       "4              2632509.99             74425.14   \n",
+       "...                   ...                  ...   \n",
+       "9723           5163900.56            587305.94   \n",
+       "9724           5361951.37            572868.36   \n",
+       "9725           6088184.20            468411.51   \n",
+       "9726           5125090.74            329522.12   \n",
+       "9727           9082482.28            492045.28   \n",
+       "\n",
+       "      Dochody_podatek_od_nieruchomosci  ...  Dochody_z_uslug   \n",
+       "0                          23823448.76  ...       2380938.91  \\\n",
+       "1                          21961009.53  ...       3748786.27   \n",
+       "2                          22107840.97  ...       4137582.84   \n",
+       "3                          22767685.01  ...       5556645.41   \n",
+       "4                          28336580.30  ...       5344170.62   \n",
+       "...                                ...  ...              ...   \n",
+       "9723                       74099232.11  ...       1859435.58   \n",
+       "9724                      108107448.79  ...       1996824.80   \n",
+       "9725                       38527846.59  ...       2187576.47   \n",
+       "9726                       78767466.83  ...       1917372.55   \n",
+       "9727                       78491368.35  ...       4110105.72   \n",
+       "\n",
+       "      Dochody_dofinansowanie_inwestycyjne  Dochody_dofinansowanie_razem   \n",
+       "0                                61031.32                    1900308.99  \\\n",
+       "1                               111299.61                     284981.40   \n",
+       "2                                89949.86                     118699.74   \n",
+       "3                                    0.00                      24676.92   \n",
+       "4                              2640028.00                    2660441.20   \n",
+       "...                                   ...                           ...   \n",
+       "9723                             20000.18                     212570.70   \n",
+       "9724                             25285.92                     237485.34   \n",
+       "9725                                 0.00                     225831.84   \n",
+       "9726                          21002107.00                   21192313.05   \n",
+       "9727                            888293.63                    1072910.83   \n",
+       "\n",
+       "      Udzialy_w_podatkach_dochodowych_od_osob_fizycznych   \n",
+       "0                                           27297399.00   \\\n",
+       "1                                           31878605.00    \n",
+       "2                                           34702412.00    \n",
+       "3                                           39128194.00    \n",
+       "4                                           40456699.00    \n",
+       "...                                                 ...    \n",
+       "9723                                        47760120.00    \n",
+       "9724                                        52799183.00    \n",
+       "9725                                        55319040.00    \n",
+       "9726                                        53739656.00    \n",
+       "9727                                        63936763.00    \n",
+       "\n",
+       "      Udzialy_w_podatkach_dochodowych_od_osob_prywatnych   \n",
+       "0                                            1984447.80   \\\n",
+       "1                                            3726450.74    \n",
+       "2                                            2536379.13    \n",
+       "3                                            1506913.18    \n",
+       "4                                            2143819.12    \n",
+       "...                                                 ...    \n",
+       "9723                                         1199589.08    \n",
+       "9724                                         2690098.17    \n",
+       "9725                                         2770684.17    \n",
+       "9726                                         3144444.38    \n",
+       "9727                                         3975531.95    \n",
+       "\n",
+       "      Udzialy_w_podatkach_dochodowych_razem  Wplywy_z_innych_lokalnych_oplat   \n",
+       "0                               29281846.80                       7018623.68  \\\n",
+       "1                               35605055.74                       7277990.62   \n",
+       "2                               37238791.13                       7339238.14   \n",
+       "3                               40635107.18                       7494838.88   \n",
+       "4                               42600518.12                       7627128.14   \n",
+       "...                                     ...                              ...   \n",
+       "9723                            48959709.08                      10930683.15   \n",
+       "9724                            55489281.17                      10458871.30   \n",
+       "9725                            58089724.17                      11369287.11   \n",
+       "9726                            56884100.38                      12281916.71   \n",
+       "9727                            67912294.95                      17127683.55   \n",
+       "\n",
+       "      Wplywy_z_oplaty_eksploatacyjnej  Wplywy_z_oplaty_skarbowej   \n",
+       "0                            28867.32                  928926.81  \\\n",
+       "1                            31466.01                  475771.76   \n",
+       "2                            24998.04                  511127.33   \n",
+       "3                            16194.57                  479489.69   \n",
+       "4                             5661.00                  440259.90   \n",
+       "...                               ...                        ...   \n",
+       "9723                         38789.78                  404331.07   \n",
+       "9724                          4684.54                  434077.88   \n",
+       "9725                          3456.95                  415686.53   \n",
+       "9726                          5157.50                  355201.29   \n",
+       "9727                         27746.70                  416473.03   \n",
+       "\n",
+       "      Wplywy_z_oplaty_targowej  \n",
+       "0                    188736.80  \n",
+       "1                    158803.00  \n",
+       "2                    131562.90  \n",
+       "3                    124935.70  \n",
+       "4                     62502.50  \n",
+       "...                        ...  \n",
+       "9723                 622509.60  \n",
+       "9724                 608625.90  \n",
+       "9725                 610059.50  \n",
+       "9726                 507341.00  \n",
+       "9727                      0.00  \n",
+       "\n",
+       "[9728 rows x 27 columns]"
+      ]
+     },
+     "execution_count": 37,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "df_data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 38,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# df_data[df_data.isna().any(axis=1)]  # ['Rok'].drop_duplicates().reset_index(drop=True)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Preprocessing"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "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": "3.11.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}