{
"metadata": {
"name": "localfinancedata_analysis"
},
"nbformat": 3,
"nbformat_minor": 0,
"worksheets": [
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
" \n",
""
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Analyse des finances des communes sur 2012\n",
"\n",
" * *source des donn\u00e9es financi\u00e8res*: [http://www.collectivites-locales.gouv.fr/](http://www.collectivites-locales.gouv.fr/)\n",
" * *source des codes communes*: [insee](http://www.insee.fr/fr/methodes/nomenclatures/cog/documentation.asp)\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"\n",
"### Examinons le nombre de communes crawl\u00e9es sur le site collectivites-locales.gouv.fr et le nombre de d\u00e9partements concern\u00e9s.\n",
"\n",
"* On s'attend \u00e0 avoir r\u00e9cup\u00e9rer les informations financi\u00e8res de 36 700 communes puisque c'est le nombre de communes d\u00e9clar\u00e9es en 2012 ([wikip\u00e9dia](http://www.insee.fr/fr/methodes/nomenclatures/cog/documentation.asp)).\n",
"* On attend \u00e9galement que ces communes soient r\u00e9parties dans 101 d\u00e9partements ([insee](http://www.insee.fr/fr/methodes/nomenclatures/cog/documentation.asp))"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"import json\n",
"import os, sys\n",
"import pandas as pd\n",
"curdir = os.path.abspath('./..')\n",
"\n",
"from localfinance.spiders.localfinance_spider import uniformize_code, convert_dom_code, convert_city\n",
"\n",
"data_dir = os.path.join(curdir, 'scraped_data')\n",
"insee_filepath = os.path.join(curdir, 'data/france2013.txt')\n",
"\n",
"quantiles = pd.np.arange(0.01, 1, 0.01)\n",
"def plot_quantiles(series, figsize=(12,10)):\n",
" pd.Series(quantiles, index=[series.quantile(q) for q in quantiles]).plot(figsize=figsize)\n",
" \n",
"from localfinance.account_parsing import city_account"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 192
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# Load insee code of cities\n",
"insee_df = pd.io.parsers.read_csv(insee_filepath, '\\t')"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 140
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# Load data on 2012\n",
"df = pd.DataFrame(json.load(open(os.path.join(data_dir, 'cities_2012.json'))))"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 142
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"print \"Les informations suivantes ont \u00e9t\u00e9 r\u00e9cup\u00e9r\u00e9es pour chacune des communes: \\n %s\"%df.columns.tolist()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"Les informations suivantes ont \u00e9t\u00e9 r\u00e9cup\u00e9r\u00e9es pour chacune des communes: \n",
" [u'additionnal_land_property_tax_basis', u'additionnal_land_property_tax_cuts_on_deliberation', u'additionnal_land_property_tax_rate', u'additionnal_land_property_tax_value', u'allocation', u'business_network_tax_value', u'business_profit_contribution_value', u'business_property_contribution_basis', u'business_property_contribution_cuts_on_deliberation', u'business_property_contribution_rate', u'business_property_contribution_value', u'contigents', u'costs_to_allocate', u'debt_annual_costs', u'debt_at_end_year', u'debt_repayment_capacity', u'debt_repayments', u'facilities_expenses', u'fctva', u'financial_costs', u'financing_capacity', u'fixed_assets', u'global_profit', u'home_tax_basis', u'home_tax_cuts_on_deliberation', u'home_tax_rate', u'home_tax_value', u'insee_code', u'investment_ressources', u'investments_usage', u'land_property_tax_basis', u'land_property_tax_cuts_on_deliberation', u'land_property_tax_rate', u'land_property_tax_value', u'loans', u'localtax', u'name', u'net_profit', u'operating_costs', u'operating_revenues', u'other_tax', u'paid_subsidies', u'population', u'property_tax_basis', u'property_tax_cuts_on_deliberation', u'property_tax_rate', u'property_tax_value', u'purchases_and_external_costs', u'received_subsidies', u'residual_financing_capacity', u'retail_land_tax_value', u'returned_properties', u'self_financing_capacity', u'staff_costs', u'surplus', u'thirdparty_balance', u'working_capital', u'year', u'zone_type']\n"
]
}
],
"prompt_number": 144
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"names_mapping = pd.DataFrame([dict([(k, v.get('name', '')) for k, v in city_account.nodes.items()])])\n",
"names_mapping.transpose()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"
\n",
"
\n",
" \n",
" \n",
" \n",
" 0 \n",
" \n",
" \n",
" \n",
" \n",
" additionnal_land_property_tax \n",
" Taxe additionnelle \u00e0 la taxe fonci\u00e8re sur les ... \n",
" \n",
" \n",
" advances_from_treasury \n",
" Avances du Tr\u00e9sor au 31/12/N \n",
" \n",
" \n",
" allocation \n",
" Dotation globale de fonctionnement \n",
" \n",
" \n",
" allocation_tax_revenues \n",
" Les produits des imp\u00f4ts de r\u00e9partition \n",
" \n",
" \n",
" business_network_tax \n",
" Impositions forfaitaires sur les entreprises d... \n",
" \n",
" \n",
" business_profit_contribution \n",
" Cotisation sur la valeur ajout\u00e9e des entreprises \n",
" \n",
" \n",
" business_property_contribution \n",
" Cotisation fonci\u00e8re des entreprises \n",
" \n",
" \n",
" business_tax \n",
" [Taxe professionnelle (hors produits \u00e9cr\u00eat\u00e9s),... \n",
" \n",
" \n",
" compensation_2010 \n",
" Compensation-Relais 2010 \n",
" \n",
" \n",
" contigents \n",
" Contingents \n",
" \n",
" \n",
" costs_to_allocate \n",
" Charges \u00e0 r\u00e9partir \n",
" \n",
" \n",
" debt_annual_costs \n",
" Annuit\u00e9 de la dette \n",
" \n",
" \n",
" debt_at_end_year \n",
" Encours total de la dette au 31/12/N \n",
" \n",
" \n",
" debt_repayment_capacity \n",
" CAF nette du remboursement en capital des empr... \n",
" \n",
" \n",
" debt_repayments \n",
" Remboursement d'emprunts et dettes assimil\u00e9es \n",
" \n",
" \n",
" facilities_expenses \n",
" D\u00e9penses d'\u00e9quipement \n",
" \n",
" \n",
" fctva \n",
" FCTVA \n",
" \n",
" \n",
" financial_costs \n",
" Charges financi\u00e8res \n",
" \n",
" \n",
" fixed_assets \n",
" Immobilisations affect\u00e9es, conc\u00e9d\u00e9es, ... \n",
" \n",
" \n",
" home_tax \n",
" [Taxe d'habitation (y compris THLV), Produits ... \n",
" \n",
" \n",
" investment_ressources \n",
" TOTAL DES RESSOURCES D'INVESTISSEMENT = C \n",
" \n",
" \n",
" investments \n",
" OPERATIONS D'INVESTISSEMENT \n",
" \n",
" \n",
" investments_usage \n",
" TOTAL DES EMPLOIS D'INVESTISSEMENT = D \n",
" \n",
" \n",
" land_property_tax \n",
" [Taxe fonci\u00e8re sur les propri\u00e9t\u00e9s non b\u00e2ties, ... \n",
" \n",
" \n",
" liabilities \n",
" ENDETTEMENT \n",
" \n",
" \n",
" loans \n",
" Emprunts bancaires et dettes assimil\u00e9es \n",
" \n",
" \n",
" localtax \n",
" Imp\u00f4ts Locaux \n",
" \n",
" \n",
" net_profit \n",
" RESULTAT COMPTABLE = A - B = R \n",
" \n",
" \n",
" operating_costs \n",
" TOTAL DES CHARGES DE FONCTIONNEMENT = B \n",
" \n",
" \n",
" operating_revenues \n",
" TOTAL DES PRODUITS DE FONCTIONNEMENT = A \n",
" \n",
" \n",
" operatings_operations \n",
" OPERATIONS DE FONCTIONNEMENT \n",
" \n",
" \n",
" other_tax \n",
" Autres imp\u00f4ts et taxes \n",
" \n",
" \n",
" paid_subsidies \n",
" Subventions vers\u00e9es \n",
" \n",
" \n",
" property_tax \n",
" [Taxe fonci\u00e8re sur les propri\u00e9t\u00e9s b\u00e2ties, Prod... \n",
" \n",
" \n",
" purchases_and_external_costs \n",
" Achats et charges externes \n",
" \n",
" \n",
" received_subsidies \n",
" Subventions re\u00e7ues \n",
" \n",
" \n",
" retail_land_tax \n",
" Taxe sur les surfaces commerciales \n",
" \n",
" \n",
" returned_properties \n",
" Retour de biens affect\u00e9s, conc\u00e9d\u00e9s, ... \n",
" \n",
" \n",
" root \n",
" \n",
" \n",
" \n",
" self_financing \n",
" AUTOFINANCEMENT \n",
" \n",
" \n",
" self_financing_capacity \n",
" Capacit\u00e9 d'autofinancement = CAF \n",
" \n",
" \n",
" staff_costs \n",
" Charges de personnel \n",
" \n",
" \n",
" surplus \n",
" Exc\u00e9dent brut de fonctionnement \n",
" \n",
" \n",
" taxation \n",
" ELEMENTS DE FISCALITE DIRECTE LOCALE \n",
" \n",
" \n",
" thirdparty_balance \n",
" Solde des op\u00e9rations pour le compte de tiers \n",
" \n",
" \n",
"
\n",
"
"
],
"output_type": "pyout",
"prompt_number": 207,
"text": [
" 0\n",
"additionnal_land_property_tax Taxe additionnelle \u00e0 la taxe fonci\u00e8re sur les ...\n",
"advances_from_treasury Avances du Tr\u00e9sor au 31/12/N\n",
"allocation Dotation globale de fonctionnement\n",
"allocation_tax_revenues Les produits des imp\u00f4ts de r\u00e9partition\n",
"business_network_tax Impositions forfaitaires sur les entreprises d...\n",
"business_profit_contribution Cotisation sur la valeur ajout\u00e9e des entreprises\n",
"business_property_contribution Cotisation fonci\u00e8re des entreprises\n",
"business_tax [Taxe professionnelle (hors produits \u00e9cr\u00eat\u00e9s),...\n",
"compensation_2010 Compensation-Relais 2010\n",
"contigents Contingents\n",
"costs_to_allocate Charges \u00e0 r\u00e9partir\n",
"debt_annual_costs Annuit\u00e9 de la dette\n",
"debt_at_end_year Encours total de la dette au 31/12/N\n",
"debt_repayment_capacity CAF nette du remboursement en capital des empr...\n",
"debt_repayments Remboursement d'emprunts et dettes assimil\u00e9es\n",
"facilities_expenses D\u00e9penses d'\u00e9quipement\n",
"fctva FCTVA\n",
"financial_costs Charges financi\u00e8res\n",
"fixed_assets Immobilisations affect\u00e9es, conc\u00e9d\u00e9es, ...\n",
"home_tax [Taxe d'habitation (y compris THLV), Produits ...\n",
"investment_ressources TOTAL DES RESSOURCES D'INVESTISSEMENT = C\n",
"investments OPERATIONS D'INVESTISSEMENT\n",
"investments_usage TOTAL DES EMPLOIS D'INVESTISSEMENT = D\n",
"land_property_tax [Taxe fonci\u00e8re sur les propri\u00e9t\u00e9s non b\u00e2ties, ...\n",
"liabilities ENDETTEMENT\n",
"loans Emprunts bancaires et dettes assimil\u00e9es\n",
"localtax Imp\u00f4ts Locaux\n",
"net_profit RESULTAT COMPTABLE = A - B = R\n",
"operating_costs TOTAL DES CHARGES DE FONCTIONNEMENT = B\n",
"operating_revenues TOTAL DES PRODUITS DE FONCTIONNEMENT = A\n",
"operatings_operations OPERATIONS DE FONCTIONNEMENT\n",
"other_tax Autres imp\u00f4ts et taxes\n",
"paid_subsidies Subventions vers\u00e9es\n",
"property_tax [Taxe fonci\u00e8re sur les propri\u00e9t\u00e9s b\u00e2ties, Prod...\n",
"purchases_and_external_costs Achats et charges externes\n",
"received_subsidies Subventions re\u00e7ues\n",
"retail_land_tax Taxe sur les surfaces commerciales\n",
"returned_properties Retour de biens affect\u00e9s, conc\u00e9d\u00e9s, ...\n",
"root \n",
"self_financing AUTOFINANCEMENT\n",
"self_financing_capacity Capacit\u00e9 d'autofinancement = CAF\n",
"staff_costs Charges de personnel\n",
"surplus Exc\u00e9dent brut de fonctionnement\n",
"taxation ELEMENTS DE FISCALITE DIRECTE LOCALE\n",
"thirdparty_balance Solde des op\u00e9rations pour le compte de tiers"
]
}
],
"prompt_number": 207
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# Sanity checks\n",
"print u\"Nombre de communes dupliqu\u00e9es: %s\"%(df['insee_code'].count() - df['insee_code'].unique().size)\n",
"print u\"Numbre of communes non dupliqu\u00e9es: %s\"%df['insee_code'].unique().size\n",
"df['dep'] = df['insee_code'].apply(lambda r: r[:3])\n",
"gp_by_dep = df.groupby('dep')\n",
"print u\"Nombre de d\u00e9partements: %s\"%gp_by_dep.dep.size().size"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"Nombre de communes dupliqu\u00e9es: 0\n",
"Numbre of communes non dupliqu\u00e9es: 36663\n",
"Nombre de d\u00e9partements: 100"
]
},
{
"output_type": "stream",
"stream": "stdout",
"text": [
"\n"
]
}
],
"prompt_number": 170
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# Take only current cities (Cf. insee doc http://www.insee.fr/fr/methodes/default.asp?page=nomenclatures/cog/doc_ffrancee.htm)\n",
"insee_df['DEP'] = uniformize_code(insee_df, 'DEP')\n",
"insee_df['COM'] = uniformize_code(insee_df, 'COM')\n",
"insee_df['DEP'] = convert_dom_code(insee_df)\n",
"insee_df['COM'] = insee_df.apply(convert_city, axis=1)\n",
"insee_df['DEPCOM'] = insee_df['DEP'] + insee_df['COM']\n",
"# Remove MAYOTTE department, \n",
"current_insee_df = insee_df[(insee_df['ACTUAL'] == 1) & (insee_df['DEP'] <> '976')] \n",
"print \"Nombre de commune sans le d\u00e9partement de Mayotte au 2013/01/01 selon l'insee: %s\"%current_insee_df['DEPCOM'].unique().size"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"Nombre de commune sans le d\u00e9partement de Mayotte au 2013/01/01 selon l'insee: 36664\n"
]
}
],
"prompt_number": 171
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Observations\n",
" * Un d\u00e9partement n'est pas repr\u00e9sent\u00e9: MAYOTTE (976), ce qui est attendu vu que les donn\u00e9es financi\u00e8res des communes de mayotte ne semble pas \u00eatre pr\u00e9sentes sur le site crawl\u00e9. Le d\u00e9partement de Mayotte compte 17 communes.\n",
" * Le nombre de communes crawl\u00e9es est de 36663 ce qui est tr\u00e8s proche du nombre total de communes en 2012: 36700 - 17 (Mayotte). Il en manque donc 20.\n",
"Nous allons donc regarder de plus pr\u00e8s si certains budgets de communes seraient absents du site [collectivites-locales.gouv.fr](http://www.collectivites-locales.gouv.fr/)\n",
"\n",
"----\n",
"\n",
"### 20 communes ne pr\u00e9sentent aucune donn\u00e9e financi\u00e8re sur 2012\n",
"\n",
"Certaines communes ne pr\u00e9sentent aucune donn\u00e9es financi\u00e8res sur 2012 sur le site des collectivit\u00e9s locales.\n",
"\n",
"Quelques exemples de page o\u00f9 il n'y a effectivement aucune donn\u00e9e sur 2012:\n",
"\n",
" * [Charleville-M\u00e9zi\u00e8res](http://alize2.finances.gouv.fr/communes/eneuro/tableau.php?icom=276&dep=074&type=BPS¶m=0)\n",
" * [Taninge](http://alize2.finances.gouv.fr/communes/eneuro/tableau.php?icom=105&dep=008&type=BPS¶m=0)\n",
" * [Grandville](http://alize2.finances.gouv.fr/communes/eneuro/tableau.php?icom=199&dep=008&type=BPS¶m=0)\n",
"\n",
"On notera que les donn\u00e9es sont bien pr\u00e9sentes sur les ann\u00e9es pr\u00e9c\u00e9dentes.\n",
"\n",
"\n",
"\n",
"Ci-dessous les 20 communes sans donn\u00e9e financi\u00e8res.\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"---\n",
"### R\u00e9sum\u00e9 statistiques des principales lignes du budget des communes\n",
"(moyenne, \u00e9cart-type, min, quantile 25%, 50%, 75% et max)\n",
"\n",
"On notera que les charges de personnels (staff_costs) peuvent \u00eatre n\u00e9gatives!\n",
"\n",
" * Exemple de la commune [Villiers-sur-Suize](http://alize2.finances.gouv.fr/communes/eneuro/detail.php?icom=538&dep=052&type=BPS¶m=0&exercice=2012)"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"df[['operating_revenues', 'operating_costs', 'staff_costs', 'net_profit', 'investment_ressources',\n",
" 'financing_capacity', 'investments_usage', 'debt_at_end_year', 'debt_annual_costs']].describe()[1:]"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" \n",
" operating_revenues \n",
" operating_costs \n",
" staff_costs \n",
" net_profit \n",
" investment_ressources \n",
" financing_capacity \n",
" investments_usage \n",
" debt_at_end_year \n",
" debt_annual_costs \n",
" \n",
" \n",
" \n",
" \n",
" mean \n",
" 2.200283e+06 \n",
" 1.922053e+06 \n",
" 9.267096e+05 \n",
" 2.782336e+05 \n",
" 9.567364e+05 \n",
" -24645.473638 \n",
" 9.324207e+05 \n",
" 1.671028e+06 \n",
" 2.119362e+05 \n",
" \n",
" \n",
" std \n",
" 3.027319e+07 \n",
" 2.792053e+07 \n",
" 1.165512e+07 \n",
" 2.435761e+06 \n",
" 1.402173e+07 \n",
" 781739.813490 \n",
" 1.367754e+07 \n",
" 2.179044e+07 \n",
" 2.386009e+06 \n",
" \n",
" \n",
" min \n",
" 9.000000e+03 \n",
" 5.000000e+03 \n",
" -1.300000e+04 \n",
" -2.968000e+06 \n",
" 0.000000e+00 \n",
" -60924000.000000 \n",
" 0.000000e+00 \n",
" -2.800000e+04 \n",
" 0.000000e+00 \n",
" \n",
" \n",
" 25% \n",
" 1.410000e+05 \n",
" 1.030000e+05 \n",
" 2.600000e+04 \n",
" 2.800000e+04 \n",
" 4.500000e+04 \n",
" -53000.000000 \n",
" 4.800000e+04 \n",
" 3.900000e+04 \n",
" 7.000000e+03 \n",
" \n",
" \n",
" 50% \n",
" 3.080000e+05 \n",
" 2.360000e+05 \n",
" 8.100000e+04 \n",
" 6.600000e+04 \n",
" 1.350000e+05 \n",
" 1000.000000 \n",
" 1.360000e+05 \n",
" 1.710000e+05 \n",
" 2.500000e+04 \n",
" \n",
" \n",
" 75% \n",
" 8.110000e+05 \n",
" 6.320000e+05 \n",
" 2.560000e+05 \n",
" 1.720000e+05 \n",
" 4.270000e+05 \n",
" 47000.000000 \n",
" 4.150000e+05 \n",
" 6.025000e+05 \n",
" 8.300000e+04 \n",
" \n",
" \n",
" max \n",
" 5.218325e+09 \n",
" 4.820358e+09 \n",
" 1.878621e+09 \n",
" 3.979670e+08 \n",
" 2.450566e+09 \n",
" 18801000.000000 \n",
" 2.389281e+09 \n",
" 3.260027e+09 \n",
" 2.806090e+08 \n",
" \n",
" \n",
"
\n",
"
"
],
"output_type": "pyout",
"prompt_number": 167,
"text": [
" operating_revenues operating_costs staff_costs net_profit \\\n",
"mean 2.200283e+06 1.922053e+06 9.267096e+05 2.782336e+05 \n",
"std 3.027319e+07 2.792053e+07 1.165512e+07 2.435761e+06 \n",
"min 9.000000e+03 5.000000e+03 -1.300000e+04 -2.968000e+06 \n",
"25% 1.410000e+05 1.030000e+05 2.600000e+04 2.800000e+04 \n",
"50% 3.080000e+05 2.360000e+05 8.100000e+04 6.600000e+04 \n",
"75% 8.110000e+05 6.320000e+05 2.560000e+05 1.720000e+05 \n",
"max 5.218325e+09 4.820358e+09 1.878621e+09 3.979670e+08 \n",
"\n",
" investment_ressources financing_capacity investments_usage \\\n",
"mean 9.567364e+05 -24645.473638 9.324207e+05 \n",
"std 1.402173e+07 781739.813490 1.367754e+07 \n",
"min 0.000000e+00 -60924000.000000 0.000000e+00 \n",
"25% 4.500000e+04 -53000.000000 4.800000e+04 \n",
"50% 1.350000e+05 1000.000000 1.360000e+05 \n",
"75% 4.270000e+05 47000.000000 4.150000e+05 \n",
"max 2.450566e+09 18801000.000000 2.389281e+09 \n",
"\n",
" debt_at_end_year debt_annual_costs \n",
"mean 1.671028e+06 2.119362e+05 \n",
"std 2.179044e+07 2.386009e+06 \n",
"min -2.800000e+04 0.000000e+00 \n",
"25% 3.900000e+04 7.000000e+03 \n",
"50% 1.710000e+05 2.500000e+04 \n",
"75% 6.025000e+05 8.300000e+04 \n",
"max 3.260027e+09 2.806090e+08 "
]
}
],
"prompt_number": 167
},
{
"cell_type": "code",
"collapsed": false,
"input": [
", 'property_tax_rate', 'land_property_tax_rate',\n",
" 'additionnal_land_property_tax_rate', 'business_property_contribution_rate']].describe()[1:]"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" \n",
" home_tax_rate \n",
" property_tax_rate \n",
" land_property_tax_rate \n",
" additionnal_land_property_tax_rate \n",
" business_property_contribution_rate \n",
" \n",
" \n",
" \n",
" \n",
" mean \n",
" 0.131094 \n",
" 0.139850 \n",
" 0.447765 \n",
" 0.221133 \n",
" 0.085261 \n",
" \n",
" \n",
" std \n",
" 0.050662 \n",
" 0.066394 \n",
" 0.282152 \n",
" 0.274089 \n",
" 0.097964 \n",
" \n",
" \n",
" min \n",
" 0.000000 \n",
" 0.000000 \n",
" 0.000000 \n",
" 0.000000 \n",
" 0.000000 \n",
" \n",
" \n",
" 25% \n",
" 0.094700 \n",
" 0.093000 \n",
" 0.260500 \n",
" 0.000000 \n",
" 0.000000 \n",
" \n",
" \n",
" 50% \n",
" 0.127400 \n",
" 0.132500 \n",
" 0.391900 \n",
" 0.000000 \n",
" 0.000000 \n",
" \n",
" \n",
" 75% \n",
" 0.162900 \n",
" 0.177600 \n",
" 0.562400 \n",
" 0.376000 \n",
" 0.171500 \n",
" \n",
" \n",
" max \n",
" 0.610300 \n",
" 0.589400 \n",
" 3.744500 \n",
" 1.369600 \n",
" 0.505700 \n",
" \n",
" \n",
"
\n",
"
"
],
"output_type": "pyout",
"prompt_number": 162,
"text": [
" home_tax_rate property_tax_rate land_property_tax_rate \\\n",
"mean 0.131094 0.139850 0.447765 \n",
"std 0.050662 0.066394 0.282152 \n",
"min 0.000000 0.000000 0.000000 \n",
"25% 0.094700 0.093000 0.260500 \n",
"50% 0.127400 0.132500 0.391900 \n",
"75% 0.162900 0.177600 0.562400 \n",
"max 0.610300 0.589400 3.744500 \n",
"\n",
" additionnal_land_property_tax_rate business_property_contribution_rate \n",
"mean 0.221133 0.085261 \n",
"std 0.274089 0.097964 \n",
"min 0.000000 0.000000 \n",
"25% 0.000000 0.000000 \n",
"50% 0.000000 0.000000 \n",
"75% 0.376000 0.171500 \n",
"max 1.369600 0.505700 "
]
}
],
"prompt_number": 162
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Analyse des taxes d'habitation, taxes fonci\u00e8res, charges de personnel et ratio d'endettement\n",
"\n",
"On s'int\u00e9resse dans cette partie \u00e0 la qualit\u00e9 de quelques informations financi\u00e8res: on commence par des informations relativement bien connues du grand public, et donc simple \u00e0 appr\u00e9hender."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"codes = set(current_insee_df['DEPCOM']).symmetric_difference(df['insee_code']).intersection(current_insee_df['DEPCOM'])\n",
"com_nodata = current_insee_df[['DEPCOM', 'NCC', 'DEP']][current_insee_df['DEPCOM'].apply(lambda r: r in codes)]\n",
"com_nodata.head(n=30)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" \n",
" DEPCOM \n",
" NCC \n",
" DEP \n",
" \n",
" \n",
" \n",
" \n",
" 3291 \n",
" 006121 \n",
" SAINT-JEAN-CAP-FERRAT \n",
" 006 \n",
" \n",
" \n",
" 5001 \n",
" 008105 \n",
" CHARLEVILLE-MEZIERES \n",
" 008 \n",
" \n",
" \n",
" 5330 \n",
" 008199 \n",
" GRANDVILLE \n",
" 008 \n",
" \n",
" \n",
" 5929 \n",
" 028285 \n",
" OINVILLE-SOUS-AUNEAU \n",
" 028 \n",
" \n",
" \n",
" 6974 \n",
" 025226 \n",
" ETRAPPE \n",
" 025 \n",
" \n",
" \n",
" 9066 \n",
" 023096 \n",
" GUERET \n",
" 023 \n",
" \n",
" \n",
" 9908 \n",
" 027226 \n",
" ETREPAGNY \n",
" 027 \n",
" \n",
" \n",
" 10251 \n",
" 027284 \n",
" GISORS \n",
" 027 \n",
" \n",
" \n",
" 10267 \n",
" 027304 \n",
" GUERNY \n",
" 027 \n",
" \n",
" \n",
" 13372 \n",
" 038254 \n",
" MONTEYNARD \n",
" 038 \n",
" \n",
" \n",
" 13383 \n",
" 038269 \n",
" MURE \n",
" 038 \n",
" \n",
" \n",
" 16155 \n",
" 051535 \n",
" SEZANNE \n",
" 051 \n",
" \n",
" \n",
" 25492 \n",
" 074273 \n",
" SIXT-FER-A-CHEVAL \n",
" 074 \n",
" \n",
" \n",
" 25494 \n",
" 074276 \n",
" TANINGES \n",
" 074 \n",
" \n",
" \n",
" 25507 \n",
" 074294 \n",
" VERCHAIX \n",
" 074 \n",
" \n",
" \n",
" 32808 \n",
" 089460 \n",
" VILLENEUVE-LA-GUYARD \n",
" 089 \n",
" \n",
" \n",
" 33750 \n",
" 006159 \n",
" VILLEFRANCHE-SUR-MER \n",
" 006 \n",
" \n",
" \n",
" 35506 \n",
" 080188 \n",
" CHAUSSOY-EPAGNY \n",
" 080 \n",
" \n",
" \n",
" 36311 \n",
" 080387 \n",
" GRATTEPANCHE \n",
" 080 \n",
" \n",
" \n",
" 37682 \n",
" 101117 \n",
" MOULE \n",
" 101 \n",
" \n",
" \n",
"
\n",
"
"
],
"output_type": "pyout",
"prompt_number": 174,
"text": [
" DEPCOM NCC DEP\n",
"3291 006121 SAINT-JEAN-CAP-FERRAT 006\n",
"5001 008105 CHARLEVILLE-MEZIERES 008\n",
"5330 008199 GRANDVILLE 008\n",
"5929 028285 OINVILLE-SOUS-AUNEAU 028\n",
"6974 025226 ETRAPPE 025\n",
"9066 023096 GUERET 023\n",
"9908 027226 ETREPAGNY 027\n",
"10251 027284 GISORS 027\n",
"10267 027304 GUERNY 027\n",
"13372 038254 MONTEYNARD 038\n",
"13383 038269 MURE 038\n",
"16155 051535 SEZANNE 051\n",
"25492 074273 SIXT-FER-A-CHEVAL 074\n",
"25494 074276 TANINGES 074\n",
"25507 074294 VERCHAIX 074\n",
"32808 089460 VILLENEUVE-LA-GUYARD 089\n",
"33750 006159 VILLEFRANCHE-SUR-MER 006\n",
"35506 080188 CHAUSSOY-EPAGNY 080\n",
"36311 080387 GRATTEPANCHE 080\n",
"37682 101117 MOULE 101"
]
}
],
"prompt_number": 174
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"---"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# Prepare data for analysis\n",
"df['property_tax_value_per_person'] = df['property_tax_value']/df['population']\n",
"df['home_tax_value_per_person'] = df['home_tax_value']/df['population']\n",
"df['debt_ratio'] = df['debt_annual_costs']/df['operating_revenues']\n",
"df['staff_costs_ratio'] = df['staff_costs']/df['operating_revenues']\n",
"df['staff_costs_per_person'] = df['staff_costs']/df['population']"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 168
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### R\u00e9partition de la taxe d'habitation et de la taxe fonci\u00e8re\n",
"\n",
"On note que la taxe fonci\u00e8re et la taxe d'habitation sont en moyenne tr\u00e8s proche, de l'ordre de 13%, avec un \u00e9cart-type un peu plus important sur la taxe fonci\u00e8re."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"df[['property_tax_rate', 'home_tax_rate']].describe()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" \n",
" property_tax_rate \n",
" home_tax_rate \n",
" \n",
" \n",
" \n",
" \n",
" count \n",
" 36663.000000 \n",
" 36663.000000 \n",
" \n",
" \n",
" mean \n",
" 0.139850 \n",
" 0.131094 \n",
" \n",
" \n",
" std \n",
" 0.066394 \n",
" 0.050662 \n",
" \n",
" \n",
" min \n",
" 0.000000 \n",
" 0.000000 \n",
" \n",
" \n",
" 25% \n",
" 0.093000 \n",
" 0.094700 \n",
" \n",
" \n",
" 50% \n",
" 0.132500 \n",
" 0.127400 \n",
" \n",
" \n",
" 75% \n",
" 0.177600 \n",
" 0.162900 \n",
" \n",
" \n",
" max \n",
" 0.589400 \n",
" 0.610300 \n",
" \n",
" \n",
"
\n",
"
"
],
"output_type": "pyout",
"prompt_number": 175,
"text": [
" property_tax_rate home_tax_rate\n",
"count 36663.000000 36663.000000\n",
"mean 0.139850 0.131094\n",
"std 0.066394 0.050662\n",
"min 0.000000 0.000000\n",
"25% 0.093000 0.094700\n",
"50% 0.132500 0.127400\n",
"75% 0.177600 0.162900\n",
"max 0.589400 0.610300"
]
}
],
"prompt_number": 175
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Une disparit\u00e9 entre les communes qui peut \u00eatre tr\u00e8s importante: on passe de 0% \u00e0 59% pour la taxe fonci\u00e8re, de 0% \u00e0 61% pour la taxe d'habitation."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"plt.figure(figsize=(12,12));\n",
"df[['property_tax_rate', 'home_tax_rate']].boxplot()\n",
"df[['property_tax_rate', 'home_tax_rate', 'name', 'insee_code']].head(n=20)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" \n",
" property_tax_rate \n",
" home_tax_rate \n",
" name \n",
" insee_code \n",
" \n",
" \n",
" \n",
" \n",
" 0 \n",
" 0.2198 \n",
" 0.1382 \n",
" ROQUEFORT-SUR-SOULZON \n",
" 012203 \n",
" \n",
" \n",
" 1 \n",
" 0.1910 \n",
" 0.1273 \n",
" ROQUE-SAINTE-MARGUERITE (LA) \n",
" 012204 \n",
" \n",
" \n",
" 2 \n",
" 0.0998 \n",
" 0.1154 \n",
" ROUSSENNAC \n",
" 012206 \n",
" \n",
" \n",
" 3 \n",
" 0.0579 \n",
" 0.0715 \n",
" RULHAC-SAINT-CIRQ \n",
" 012207 \n",
" \n",
" \n",
" 4 \n",
" 0.2670 \n",
" 0.1145 \n",
" SAINT-AFFRIQUE \n",
" 012208 \n",
" \n",
" \n",
" 5 \n",
" 0.0622 \n",
" 0.0308 \n",
" SAINT-AMANS-DES-COTS \n",
" 012209 \n",
" \n",
" \n",
" 6 \n",
" 0.2100 \n",
" 0.1678 \n",
" SAINT-ANDRE-DE-NAJAC \n",
" 012210 \n",
" \n",
" \n",
" 7 \n",
" 0.0568 \n",
" 0.0451 \n",
" SAINT-BEAULIZE \n",
" 012212 \n",
" \n",
" \n",
" 8 \n",
" 0.1277 \n",
" 0.0970 \n",
" SAINT-ANDRE-DE-VEZINES \n",
" 012211 \n",
" \n",
" \n",
" 9 \n",
" 0.1917 \n",
" 0.1265 \n",
" SAINT-CHELY-D'AUBRAC \n",
" 012214 \n",
" \n",
" \n",
" 10 \n",
" 0.1582 \n",
" 0.1394 \n",
" SAINT-BEAUZELY \n",
" 012213 \n",
" \n",
" \n",
" 11 \n",
" 0.1425 \n",
" 0.0766 \n",
" SAINT-CHRISTOPHE-VALLON \n",
" 012215 \n",
" \n",
" \n",
" 12 \n",
" 0.1236 \n",
" 0.1395 \n",
" SAINT-COME-D'OLT \n",
" 012216 \n",
" \n",
" \n",
" 13 \n",
" 0.1339 \n",
" 0.0825 \n",
" SAINTE-CROIX \n",
" 012217 \n",
" \n",
" \n",
" 14 \n",
" 0.1550 \n",
" 0.0880 \n",
" SAINT-CYPRIEN-SUR-DOURDOU \n",
" 012218 \n",
" \n",
" \n",
" 15 \n",
" 0.0564 \n",
" 0.0637 \n",
" SAINTE-EULALIE-D'OLT \n",
" 012219 \n",
" \n",
" \n",
" 16 \n",
" 0.1437 \n",
" 0.1000 \n",
" SAINT-FELIX-DE-LUNEL \n",
" 012221 \n",
" \n",
" \n",
" 17 \n",
" 0.0914 \n",
" 0.1063 \n",
" SAINTE-GENEVIEVE-SUR-ARGENCE \n",
" 012223 \n",
" \n",
" \n",
" 18 \n",
" 0.1104 \n",
" 0.0635 \n",
" SAINT-FELIX-DE-SORGUES \n",
" 012222 \n",
" \n",
" \n",
" 19 \n",
" 0.1062 \n",
" 0.0628 \n",
" SAINTE-EULALIE-DE-CERNON \n",
" 012220 \n",
" \n",
" \n",
"
\n",
"
"
],
"output_type": "pyout",
"prompt_number": 176,
"text": [
" property_tax_rate home_tax_rate name insee_code\n",
"0 0.2198 0.1382 ROQUEFORT-SUR-SOULZON 012203\n",
"1 0.1910 0.1273 ROQUE-SAINTE-MARGUERITE (LA) 012204\n",
"2 0.0998 0.1154 ROUSSENNAC 012206\n",
"3 0.0579 0.0715 RULHAC-SAINT-CIRQ 012207\n",
"4 0.2670 0.1145 SAINT-AFFRIQUE 012208\n",
"5 0.0622 0.0308 SAINT-AMANS-DES-COTS 012209\n",
"6 0.2100 0.1678 SAINT-ANDRE-DE-NAJAC 012210\n",
"7 0.0568 0.0451 SAINT-BEAULIZE 012212\n",
"8 0.1277 0.0970 SAINT-ANDRE-DE-VEZINES 012211\n",
"9 0.1917 0.1265 SAINT-CHELY-D'AUBRAC 012214\n",
"10 0.1582 0.1394 SAINT-BEAUZELY 012213\n",
"11 0.1425 0.0766 SAINT-CHRISTOPHE-VALLON 012215\n",
"12 0.1236 0.1395 SAINT-COME-D'OLT 012216\n",
"13 0.1339 0.0825 SAINTE-CROIX 012217\n",
"14 0.1550 0.0880 SAINT-CYPRIEN-SUR-DOURDOU 012218\n",
"15 0.0564 0.0637 SAINTE-EULALIE-D'OLT 012219\n",
"16 0.1437 0.1000 SAINT-FELIX-DE-LUNEL 012221\n",
"17 0.0914 0.1063 SAINTE-GENEVIEVE-SUR-ARGENCE 012223\n",
"18 0.1104 0.0635 SAINT-FELIX-DE-SORGUES 012222\n",
"19 0.1062 0.0628 SAINTE-EULALIE-DE-CERNON 012220"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAtQAAAKpCAYAAACckbvvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSKgBAACAGpBQAwAAADX4AYlvt7IM3xlYAAAAAElFTkSuQmCC\n"
}
],
"prompt_number": 176
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# Biggest property tax rate\n",
"_df = df.sort(columns='property_tax_rate', ascending=False)\n",
"_df[['property_tax_rate', 'property_tax_value', 'name', 'insee_code']].head(n=20)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" \n",
" property_tax_rate \n",
" property_tax_value \n",
" name \n",
" insee_code \n",
" \n",
" \n",
" \n",
" \n",
" 26903 \n",
" 0.5894 \n",
" 18219000 \n",
" BUSSY-SAINT-GEORGES \n",
" 077058 \n",
" \n",
" \n",
" 3823 \n",
" 0.5791 \n",
" 27000 \n",
" FONTANES-DE-SAULT \n",
" 011147 \n",
" \n",
" \n",
" 16143 \n",
" 0.5658 \n",
" 42000 \n",
" CHAMBORD \n",
" 041034 \n",
" \n",
" \n",
" 32240 \n",
" 0.5645 \n",
" 236000 \n",
" MIZOEN \n",
" 038237 \n",
" \n",
" \n",
" 35052 \n",
" 0.5496 \n",
" 14305000 \n",
" SAINT-LOUIS \n",
" 104414 \n",
" \n",
" \n",
" 3874 \n",
" 0.5284 \n",
" 127000 \n",
" MAILHAC \n",
" 011212 \n",
" \n",
" \n",
" 27699 \n",
" 0.5130 \n",
" 219000 \n",
" ETOILE (L' ) \n",
" 080296 \n",
" \n",
" \n",
" 19839 \n",
" 0.5028 \n",
" 2190000 \n",
" CONDE-SUR-ESCAUT \n",
" 059153 \n",
" \n",
" \n",
" 28283 \n",
" 0.5000 \n",
" 1190000 \n",
" SAINT-COLOMBAN-DES-VILLARDS \n",
" 073230 \n",
" \n",
" \n",
" 36017 \n",
" 0.4991 \n",
" 10723000 \n",
" WATTRELOS \n",
" 059650 \n",
" \n",
" \n",
" 21116 \n",
" 0.4989 \n",
" 1710000 \n",
" SAINS-EN-GOHELLE \n",
" 062737 \n",
" \n",
" \n",
" 29417 \n",
" 0.4936 \n",
" 3728000 \n",
" COULOUNIEIX-CHAMIERS \n",
" 024138 \n",
" \n",
" \n",
" 14261 \n",
" 0.4891 \n",
" 1127000 \n",
" SAINT-SEURIN-SUR-L'ISLE \n",
" 033478 \n",
" \n",
" \n",
" 3512 \n",
" 0.4824 \n",
" 485000 \n",
" CAUNES-MINERVOIS \n",
" 011081 \n",
" \n",
" \n",
" 33686 \n",
" 0.4788 \n",
" 17000 \n",
" REMECOURT \n",
" 060529 \n",
" \n",
" \n",
" 33918 \n",
" 0.4710 \n",
" 13209000 \n",
" SAVIGNY-LE-TEMPLE \n",
" 077445 \n",
" \n",
" \n",
" 20355 \n",
" 0.4677 \n",
" 9316000 \n",
" LIEVIN \n",
" 062510 \n",
" \n",
" \n",
" 24497 \n",
" 0.4673 \n",
" 796000 \n",
" ARQUES-LA-BATAILLE \n",
" 076026 \n",
" \n",
" \n",
" 20291 \n",
" 0.4668 \n",
" 11623000 \n",
" HENIN-BEAUMONT \n",
" 062427 \n",
" \n",
" \n",
" 4162 \n",
" 0.4650 \n",
" 758000 \n",
" PENNAUTIER \n",
" 011279 \n",
" \n",
" \n",
"
\n",
"
"
],
"output_type": "pyout",
"prompt_number": 177,
"text": [
" property_tax_rate property_tax_value name insee_code\n",
"26903 0.5894 18219000 BUSSY-SAINT-GEORGES 077058\n",
"3823 0.5791 27000 FONTANES-DE-SAULT 011147\n",
"16143 0.5658 42000 CHAMBORD 041034\n",
"32240 0.5645 236000 MIZOEN 038237\n",
"35052 0.5496 14305000 SAINT-LOUIS 104414\n",
"3874 0.5284 127000 MAILHAC 011212\n",
"27699 0.5130 219000 ETOILE (L' ) 080296\n",
"19839 0.5028 2190000 CONDE-SUR-ESCAUT 059153\n",
"28283 0.5000 1190000 SAINT-COLOMBAN-DES-VILLARDS 073230\n",
"36017 0.4991 10723000 WATTRELOS 059650\n",
"21116 0.4989 1710000 SAINS-EN-GOHELLE 062737\n",
"29417 0.4936 3728000 COULOUNIEIX-CHAMIERS 024138\n",
"14261 0.4891 1127000 SAINT-SEURIN-SUR-L'ISLE 033478\n",
"3512 0.4824 485000 CAUNES-MINERVOIS 011081\n",
"33686 0.4788 17000 REMECOURT 060529\n",
"33918 0.4710 13209000 SAVIGNY-LE-TEMPLE 077445\n",
"20355 0.4677 9316000 LIEVIN 062510\n",
"24497 0.4673 796000 ARQUES-LA-BATAILLE 076026\n",
"20291 0.4668 11623000 HENIN-BEAUMONT 062427\n",
"4162 0.4650 758000 PENNAUTIER 011279"
]
}
],
"prompt_number": 177
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# Lowest property tax\n",
"_df = df.sort(columns='property_tax_rate', ascending=True)\n",
"_df[['property_tax_rate', 'name', 'insee_code']].head(n=20)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" \n",
" property_tax_rate \n",
" name \n",
" insee_code \n",
" \n",
" \n",
" \n",
" \n",
" 1871 \n",
" 0 \n",
" SUZAN \n",
" 009304 \n",
" \n",
" \n",
" 35405 \n",
" 0 \n",
" SAINT-CIRICE \n",
" 082158 \n",
" \n",
" \n",
" 6353 \n",
" 0 \n",
" ILE-DE-SEIN \n",
" 029083 \n",
" \n",
" \n",
" 1840 \n",
" 0 \n",
" MONTJUSTIN \n",
" 004129 \n",
" \n",
" \n",
" 20165 \n",
" 0 \n",
" BLANZEE \n",
" 055055 \n",
" \n",
" \n",
" 25300 \n",
" 0 \n",
" SISTELS \n",
" 082181 \n",
" \n",
" \n",
" 36324 \n",
" 0 \n",
" BRULLEMAIL \n",
" 061064 \n",
" \n",
" \n",
" 17365 \n",
" 0 \n",
" BOISSEI-LA-LANDE \n",
" 061049 \n",
" \n",
" \n",
" 36223 \n",
" 0 \n",
" SIVRY-LA-PERCHE \n",
" 055489 \n",
" \n",
" \n",
" 36596 \n",
" 0 \n",
" CHATEAU-D'ALMENECHES (LE ) \n",
" 061101 \n",
" \n",
" \n",
" 20926 \n",
" 0 \n",
" HAUMONT-PRES-SAMOGNEUX \n",
" 055239 \n",
" \n",
" \n",
" 36362 \n",
" 0 \n",
" TILLY-SUR-MEUSE \n",
" 055512 \n",
" \n",
" \n",
" 20965 \n",
" 0 \n",
" LOISON \n",
" 055299 \n",
" \n",
" \n",
" 15525 \n",
" 0 \n",
" AVRECOURT \n",
" 052033 \n",
" \n",
" \n",
" 11843 \n",
" 0 \n",
" FERRIERES \n",
" 050179 \n",
" \n",
" \n",
" 35772 \n",
" 0 \n",
" LOUVEMONT-COTE-DU-POIVRE \n",
" 055307 \n",
" \n",
" \n",
" 21740 \n",
" 0 \n",
" SENONCOURT-LES-MAUJOUY \n",
" 055482 \n",
" \n",
" \n",
" 24962 \n",
" 0 \n",
" CASTELSAGRAT \n",
" 082032 \n",
" \n",
" \n",
" 20160 \n",
" 0 \n",
" BETHELAINVILLE \n",
" 055047 \n",
" \n",
" \n",
" 20162 \n",
" 0 \n",
" BEZONVAUX \n",
" 055050 \n",
" \n",
" \n",
"
\n",
"
"
],
"output_type": "pyout",
"prompt_number": 178,
"text": [
" property_tax_rate name insee_code\n",
"1871 0 SUZAN 009304\n",
"35405 0 SAINT-CIRICE 082158\n",
"6353 0 ILE-DE-SEIN 029083\n",
"1840 0 MONTJUSTIN 004129\n",
"20165 0 BLANZEE 055055\n",
"25300 0 SISTELS 082181\n",
"36324 0 BRULLEMAIL 061064\n",
"17365 0 BOISSEI-LA-LANDE 061049\n",
"36223 0 SIVRY-LA-PERCHE 055489\n",
"36596 0 CHATEAU-D'ALMENECHES (LE ) 061101\n",
"20926 0 HAUMONT-PRES-SAMOGNEUX 055239\n",
"36362 0 TILLY-SUR-MEUSE 055512\n",
"20965 0 LOISON 055299\n",
"15525 0 AVRECOURT 052033\n",
"11843 0 FERRIERES 050179\n",
"35772 0 LOUVEMONT-COTE-DU-POIVRE 055307\n",
"21740 0 SENONCOURT-LES-MAUJOUY 055482\n",
"24962 0 CASTELSAGRAT 082032\n",
"20160 0 BETHELAINVILLE 055047\n",
"20162 0 BEZONVAUX 055050"
]
}
],
"prompt_number": 178
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# Biggest home tax rate\n",
"_df = df.sort(columns='home_tax_rate', ascending=False)\n",
"_df[['home_tax_rate', 'home_tax_value', 'name', 'insee_code']].head(n=20)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" \n",
" home_tax_rate \n",
" home_tax_value \n",
" name \n",
" insee_code \n",
" \n",
" \n",
" \n",
" \n",
" 7851 \n",
" 0.6103 \n",
" 32000 \n",
" VERDESE \n",
" 02B344 \n",
" \n",
" \n",
" 3874 \n",
" 0.5130 \n",
" 197000 \n",
" MAILHAC \n",
" 011212 \n",
" \n",
" \n",
" 31870 \n",
" 0.4891 \n",
" 365000 \n",
" SANTA-MARIA-POGGIO \n",
" 02B311 \n",
" \n",
" \n",
" 32571 \n",
" 0.4664 \n",
" 478000 \n",
" ROURA \n",
" 102310 \n",
" \n",
" \n",
" 19839 \n",
" 0.4506 \n",
" 2113000 \n",
" CONDE-SUR-ESCAUT \n",
" 059153 \n",
" \n",
" \n",
" 7132 \n",
" 0.4336 \n",
" 62000 \n",
" TASSO \n",
" 02A322 \n",
" \n",
" \n",
" 35052 \n",
" 0.4311 \n",
" 9504000 \n",
" SAINT-LOUIS \n",
" 104414 \n",
" \n",
" \n",
" 30164 \n",
" 0.4305 \n",
" 18000 \n",
" BREMONDANS \n",
" 025089 \n",
" \n",
" \n",
" 26903 \n",
" 0.4227 \n",
" 12506000 \n",
" BUSSY-SAINT-GEORGES \n",
" 077058 \n",
" \n",
" \n",
" 7452 \n",
" 0.4160 \n",
" 94000 \n",
" CORSCIA \n",
" 02B095 \n",
" \n",
" \n",
" 18321 \n",
" 0.4081 \n",
" 65000 \n",
" GAVARNIE \n",
" 065188 \n",
" \n",
" \n",
" 30022 \n",
" 0.4074 \n",
" 327000 \n",
" BOUZY \n",
" 051079 \n",
" \n",
" \n",
" 20589 \n",
" 0.4059 \n",
" 9930000 \n",
" LAMBERSART \n",
" 059328 \n",
" \n",
" \n",
" 35014 \n",
" 0.3937 \n",
" 386000 \n",
" TERRE DE HAUT \n",
" 101131 \n",
" \n",
" \n",
" 21030 \n",
" 0.3921 \n",
" 3303000 \n",
" SAINT-ANDRE-LEZ-LILLE \n",
" 059527 \n",
" \n",
" \n",
" 21741 \n",
" 0.3828 \n",
" 12000 \n",
" SEPTSARGES \n",
" 055484 \n",
" \n",
" \n",
" 7783 \n",
" 0.3777 \n",
" 50000 \n",
" PIEDICROCE \n",
" 02B219 \n",
" \n",
" \n",
" 20625 \n",
" 0.3762 \n",
" 1013000 \n",
" MARCHIENNES \n",
" 059375 \n",
" \n",
" \n",
" 20233 \n",
" 0.3740 \n",
" 1447000 \n",
" FRESNES-SUR-ESCAUT \n",
" 059253 \n",
" \n",
" \n",
" 20996 \n",
" 0.3689 \n",
" 1076000 \n",
" QUESNOY (LE ) \n",
" 059481 \n",
" \n",
" \n",
"
\n",
"
"
],
"output_type": "pyout",
"prompt_number": 179,
"text": [
" home_tax_rate home_tax_value name insee_code\n",
"7851 0.6103 32000 VERDESE 02B344\n",
"3874 0.5130 197000 MAILHAC 011212\n",
"31870 0.4891 365000 SANTA-MARIA-POGGIO 02B311\n",
"32571 0.4664 478000 ROURA 102310\n",
"19839 0.4506 2113000 CONDE-SUR-ESCAUT 059153\n",
"7132 0.4336 62000 TASSO 02A322\n",
"35052 0.4311 9504000 SAINT-LOUIS 104414\n",
"30164 0.4305 18000 BREMONDANS 025089\n",
"26903 0.4227 12506000 BUSSY-SAINT-GEORGES 077058\n",
"7452 0.4160 94000 CORSCIA 02B095\n",
"18321 0.4081 65000 GAVARNIE 065188\n",
"30022 0.4074 327000 BOUZY 051079\n",
"20589 0.4059 9930000 LAMBERSART 059328\n",
"35014 0.3937 386000 TERRE DE HAUT 101131\n",
"21030 0.3921 3303000 SAINT-ANDRE-LEZ-LILLE 059527\n",
"21741 0.3828 12000 SEPTSARGES 055484\n",
"7783 0.3777 50000 PIEDICROCE 02B219\n",
"20625 0.3762 1013000 MARCHIENNES 059375\n",
"20233 0.3740 1447000 FRESNES-SUR-ESCAUT 059253\n",
"20996 0.3689 1076000 QUESNOY (LE ) 059481"
]
}
],
"prompt_number": 179
},
{
"cell_type": "code",
"collapsed": false,
"input": [
" #Biggest home tax value\n",
"_df = df.sort(columns='home_tax_value', ascending=False).dropna()\n",
"_df[['home_tax_rate', 'home_tax_value', 'home_tax_value_per_person', 'name', 'insee_code']].head(n=20)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" \n",
" home_tax_rate \n",
" home_tax_value \n",
" home_tax_value_per_person \n",
" name \n",
" insee_code \n",
" \n",
" \n",
" \n",
" \n",
" 24238 \n",
" 0.1338 \n",
" 700350000 \n",
" 310.166472 \n",
" PARIS \n",
" 075056 \n",
" \n",
" \n",
" 384 \n",
" 0.2723 \n",
" 247219000 \n",
" 287.831441 \n",
" MARSEILLE \n",
" 013055 \n",
" \n",
" \n",
" 35041 \n",
" 0.2130 \n",
" 158478000 \n",
" 324.764641 \n",
" LYON \n",
" 069123 \n",
" \n",
" \n",
" 31238 \n",
" 0.2133 \n",
" 126427000 \n",
" 367.029554 \n",
" NICE \n",
" 006088 \n",
" \n",
" \n",
" 29635 \n",
" 0.1587 \n",
" 88933000 \n",
" 198.779158 \n",
" TOULOUSE \n",
" 031555 \n",
" \n",
" \n",
" 12929 \n",
" 0.2404 \n",
" 88282000 \n",
" 304.284286 \n",
" NANTES \n",
" 044109 \n",
" \n",
" \n",
" 32542 \n",
" 0.2298 \n",
" 84684000 \n",
" 352.084217 \n",
" BORDEAUX \n",
" 033063 \n",
" \n",
" \n",
" 14784 \n",
" 0.2249 \n",
" 73973000 \n",
" 286.310892 \n",
" MONTPELLIER \n",
" 034172 \n",
" \n",
" \n",
" 20608 \n",
" 0.3355 \n",
" 70197000 \n",
" 300.998225 \n",
" LILLE \n",
" 059350 \n",
" \n",
" \n",
" 16037 \n",
" 0.2199 \n",
" 64701000 \n",
" 304.864085 \n",
" RENNES \n",
" 035238 \n",
" \n",
" \n",
" 2789 \n",
" 0.2788 \n",
" 64130000 \n",
" 861.441333 \n",
" CANNES \n",
" 006029 \n",
" \n",
" \n",
" 35972 \n",
" 0.2406 \n",
" 63592000 \n",
" 230.292320 \n",
" STRASBOURG \n",
" 067482 \n",
" \n",
" \n",
" 8540 \n",
" 0.2933 \n",
" 47963000 \n",
" 333.268481 \n",
" NIMES \n",
" 030189 \n",
" \n",
" \n",
" 12397 \n",
" 0.2152 \n",
" 47158000 \n",
" 298.051460 \n",
" GRENOBLE \n",
" 038185 \n",
" \n",
" \n",
" 15025 \n",
" 0.2075 \n",
" 45741000 \n",
" 247.270034 \n",
" REIMS \n",
" 051454 \n",
" \n",
" \n",
" 11332 \n",
" 0.2016 \n",
" 45307000 \n",
" 258.597170 \n",
" SAINT-ETIENNE \n",
" 042218 \n",
" \n",
" \n",
" 31727 \n",
" 0.2183 \n",
" 43129000 \n",
" 565.722681 \n",
" SAINT-MAUR-DES-FOSSES \n",
" 094068 \n",
" \n",
" \n",
" 35045 \n",
" 0.1935 \n",
" 40998000 \n",
" 244.307652 \n",
" TOULON \n",
" 083137 \n",
" \n",
" \n",
" 5435 \n",
" 0.2222 \n",
" 40075000 \n",
" 256.672196 \n",
" DIJON \n",
" 021231 \n",
" \n",
" \n",
" 27818 \n",
" 0.1837 \n",
" 38470000 \n",
" 271.740282 \n",
" LIMOGES \n",
" 087085 \n",
" \n",
" \n",
"
\n",
"
"
],
"output_type": "pyout",
"prompt_number": 180,
"text": [
" home_tax_rate home_tax_value home_tax_value_per_person \\\n",
"24238 0.1338 700350000 310.166472 \n",
"384 0.2723 247219000 287.831441 \n",
"35041 0.2130 158478000 324.764641 \n",
"31238 0.2133 126427000 367.029554 \n",
"29635 0.1587 88933000 198.779158 \n",
"12929 0.2404 88282000 304.284286 \n",
"32542 0.2298 84684000 352.084217 \n",
"14784 0.2249 73973000 286.310892 \n",
"20608 0.3355 70197000 300.998225 \n",
"16037 0.2199 64701000 304.864085 \n",
"2789 0.2788 64130000 861.441333 \n",
"35972 0.2406 63592000 230.292320 \n",
"8540 0.2933 47963000 333.268481 \n",
"12397 0.2152 47158000 298.051460 \n",
"15025 0.2075 45741000 247.270034 \n",
"11332 0.2016 45307000 258.597170 \n",
"31727 0.2183 43129000 565.722681 \n",
"35045 0.1935 40998000 244.307652 \n",
"5435 0.2222 40075000 256.672196 \n",
"27818 0.1837 38470000 271.740282 \n",
"\n",
" name insee_code \n",
"24238 PARIS 075056 \n",
"384 MARSEILLE 013055 \n",
"35041 LYON 069123 \n",
"31238 NICE 006088 \n",
"29635 TOULOUSE 031555 \n",
"12929 NANTES 044109 \n",
"32542 BORDEAUX 033063 \n",
"14784 MONTPELLIER 034172 \n",
"20608 LILLE 059350 \n",
"16037 RENNES 035238 \n",
"2789 CANNES 006029 \n",
"35972 STRASBOURG 067482 \n",
"8540 NIMES 030189 \n",
"12397 GRENOBLE 038185 \n",
"15025 REIMS 051454 \n",
"11332 SAINT-ETIENNE 042218 \n",
"31727 SAINT-MAUR-DES-FOSSES 094068 \n",
"35045 TOULON 083137 \n",
"5435 DIJON 021231 \n",
"27818 LIMOGES 087085 "
]
}
],
"prompt_number": 180
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### La taxe d'habitation par personne la plus \u00e9lev\u00e9e: 4000\u20ac/hab dans Les Angles (Gard-04)"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
" #Biggest home tax value per person\n",
"_df = df.sort(columns='home_tax_value_per_person', ascending=False).dropna()\n",
"_df[['home_tax_rate', 'home_tax_value', 'home_tax_value_per_person', 'name', 'insee_code']].head(n=20)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" \n",
" home_tax_rate \n",
" home_tax_value \n",
" home_tax_value_per_person \n",
" name \n",
" insee_code \n",
" \n",
" \n",
" \n",
" \n",
" 35541 \n",
" 0.0000 \n",
" 0 \n",
" inf \n",
" FLEURY-DEVANT-DOUAUMONT \n",
" 055189 \n",
" \n",
" \n",
" 20155 \n",
" 0.0000 \n",
" 0 \n",
" inf \n",
" BEAUMONT-EN-VERDUNOIS \n",
" 055039 \n",
" \n",
" \n",
" 35772 \n",
" 0.0000 \n",
" 0 \n",
" inf \n",
" LOUVEMONT-COTE-DU-POIVRE \n",
" 055307 \n",
" \n",
" \n",
" 20547 \n",
" 0.0000 \n",
" 0 \n",
" inf \n",
" CUMIERES-LE-MORT-HOMME \n",
" 055139 \n",
" \n",
" \n",
" 20926 \n",
" 0.0000 \n",
" 0 \n",
" inf \n",
" HAUMONT-PRES-SAMOGNEUX \n",
" 055239 \n",
" \n",
" \n",
" 20162 \n",
" 0.0000 \n",
" 0 \n",
" inf \n",
" BEZONVAUX \n",
" 055050 \n",
" \n",
" \n",
" 19151 \n",
" 0.2762 \n",
" 2284000 \n",
" 3993.006993 \n",
" ANGLES (LES) \n",
" 066004 \n",
" \n",
" \n",
" 34189 \n",
" 0.2261 \n",
" 790000 \n",
" 3062.015504 \n",
" VILLAREMBERT \n",
" 073318 \n",
" \n",
" \n",
" 12402 \n",
" 0.2355 \n",
" 3598000 \n",
" 2628.195763 \n",
" HUEZ \n",
" 038191 \n",
" \n",
" \n",
" 31252 \n",
" 0.2402 \n",
" 4149000 \n",
" 2622.629583 \n",
" THEOULE-SUR-MER \n",
" 006138 \n",
" \n",
" \n",
" 19585 \n",
" 0.2614 \n",
" 197000 \n",
" 2592.105263 \n",
" PUYVALADOR \n",
" 066154 \n",
" \n",
" \n",
" 19206 \n",
" 0.3066 \n",
" 308000 \n",
" 2444.444444 \n",
" EYNE \n",
" 066075 \n",
" \n",
" \n",
" 35678 \n",
" 0.2012 \n",
" 1655000 \n",
" 2398.550725 \n",
" RAYOL-CANADEL-SUR-MER \n",
" 083152 \n",
" \n",
" \n",
" 19164 \n",
" 0.2382 \n",
" 1943000 \n",
" 2386.977887 \n",
" BOLQUERE \n",
" 066020 \n",
" \n",
" \n",
" 28906 \n",
" 0.2963 \n",
" 2824000 \n",
" 2271.922767 \n",
" CHATEL \n",
" 074063 \n",
" \n",
" \n",
" 28241 \n",
" 0.2008 \n",
" 159000 \n",
" 2271.428571 \n",
" MONTGELLAFREY \n",
" 073167 \n",
" \n",
" \n",
" 28984 \n",
" 0.2340 \n",
" 9178000 \n",
" 2251.717370 \n",
" MEGEVE \n",
" 074173 \n",
" \n",
" \n",
" 18760 \n",
" 0.3291 \n",
" 2165000 \n",
" 2209.183673 \n",
" SAINT-LARY-SOULAN \n",
" 065388 \n",
" \n",
" \n",
" 27875 \n",
" 0.1778 \n",
" 3986000 \n",
" 2063.146998 \n",
" ALLUES (LES) \n",
" 073015 \n",
" \n",
" \n",
" 28600 \n",
" 0.1741 \n",
" 3177000 \n",
" 1933.657943 \n",
" VAL-D'ISERE \n",
" 073304 \n",
" \n",
" \n",
"
\n",
"
"
],
"output_type": "pyout",
"prompt_number": 181,
"text": [
" home_tax_rate home_tax_value home_tax_value_per_person \\\n",
"35541 0.0000 0 inf \n",
"20155 0.0000 0 inf \n",
"35772 0.0000 0 inf \n",
"20547 0.0000 0 inf \n",
"20926 0.0000 0 inf \n",
"20162 0.0000 0 inf \n",
"19151 0.2762 2284000 3993.006993 \n",
"34189 0.2261 790000 3062.015504 \n",
"12402 0.2355 3598000 2628.195763 \n",
"31252 0.2402 4149000 2622.629583 \n",
"19585 0.2614 197000 2592.105263 \n",
"19206 0.3066 308000 2444.444444 \n",
"35678 0.2012 1655000 2398.550725 \n",
"19164 0.2382 1943000 2386.977887 \n",
"28906 0.2963 2824000 2271.922767 \n",
"28241 0.2008 159000 2271.428571 \n",
"28984 0.2340 9178000 2251.717370 \n",
"18760 0.3291 2165000 2209.183673 \n",
"27875 0.1778 3986000 2063.146998 \n",
"28600 0.1741 3177000 1933.657943 \n",
"\n",
" name insee_code \n",
"35541 FLEURY-DEVANT-DOUAUMONT 055189 \n",
"20155 BEAUMONT-EN-VERDUNOIS 055039 \n",
"35772 LOUVEMONT-COTE-DU-POIVRE 055307 \n",
"20547 CUMIERES-LE-MORT-HOMME 055139 \n",
"20926 HAUMONT-PRES-SAMOGNEUX 055239 \n",
"20162 BEZONVAUX 055050 \n",
"19151 ANGLES (LES) 066004 \n",
"34189 VILLAREMBERT 073318 \n",
"12402 HUEZ 038191 \n",
"31252 THEOULE-SUR-MER 006138 \n",
"19585 PUYVALADOR 066154 \n",
"19206 EYNE 066075 \n",
"35678 RAYOL-CANADEL-SUR-MER 083152 \n",
"19164 BOLQUERE 066020 \n",
"28906 CHATEL 074063 \n",
"28241 MONTGELLAFREY 073167 \n",
"28984 MEGEVE 074173 \n",
"18760 SAINT-LARY-SOULAN 065388 \n",
"27875 ALLUES (LES) 073015 \n",
"28600 VAL-D'ISERE 073304 "
]
}
],
"prompt_number": 181
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# lowest home tax rate\n",
"_df = df.sort(columns='home_tax_value', ascending=True)\n",
"_df[['home_tax_rate', 'home_tax_value', 'name', 'insee_code']].dropna().head(n=20)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" \n",
" home_tax_rate \n",
" home_tax_value \n",
" name \n",
" insee_code \n",
" \n",
" \n",
" \n",
" \n",
" 35541 \n",
" 0.0000 \n",
" 0 \n",
" FLEURY-DEVANT-DOUAUMONT \n",
" 055189 \n",
" \n",
" \n",
" 20972 \n",
" 0.0331 \n",
" 0 \n",
" MAIZERAY \n",
" 055311 \n",
" \n",
" \n",
" 2701 \n",
" 0.0001 \n",
" 0 \n",
" SAINT-VULBAS \n",
" 001390 \n",
" \n",
" \n",
" 35998 \n",
" 0.0703 \n",
" 0 \n",
" ORNES \n",
" 055394 \n",
" \n",
" \n",
" 17681 \n",
" 0.0684 \n",
" 0 \n",
" MOLRING \n",
" 057470 \n",
" \n",
" \n",
" 3132 \n",
" 0.0000 \n",
" 0 \n",
" SOULAINES-DHUYS \n",
" 010372 \n",
" \n",
" \n",
" 3164 \n",
" 0.0000 \n",
" 0 \n",
" VILLE-AUX-BOIS (LA) \n",
" 010411 \n",
" \n",
" \n",
" 28283 \n",
" 0.0001 \n",
" 0 \n",
" SAINT-COLOMBAN-DES-VILLARDS \n",
" 073230 \n",
" \n",
" \n",
" 5925 \n",
" 0.0000 \n",
" 0 \n",
" BOUVERANS \n",
" 025085 \n",
" \n",
" \n",
" 3436 \n",
" 0.0001 \n",
" 0 \n",
" CRUAS \n",
" 007076 \n",
" \n",
" \n",
" 8450 \n",
" 0.0000 \n",
" 0 \n",
" POMMEROL \n",
" 026245 \n",
" \n",
" \n",
" 17394 \n",
" 0.0102 \n",
" 0 \n",
" CHAMP-HAUT \n",
" 061088 \n",
" \n",
" \n",
" 17365 \n",
" 0.0000 \n",
" 0 \n",
" BOISSEI-LA-LANDE \n",
" 061049 \n",
" \n",
" \n",
" 17337 \n",
" 0.0097 \n",
" 0 \n",
" AUTHIEUX-DU-PUITS (LES) \n",
" 061017 \n",
" \n",
" \n",
" 36596 \n",
" 0.0000 \n",
" 0 \n",
" CHATEAU-D'ALMENECHES (LE ) \n",
" 061101 \n",
" \n",
" \n",
" 20162 \n",
" 0.0000 \n",
" 0 \n",
" BEZONVAUX \n",
" 055050 \n",
" \n",
" \n",
" 10708 \n",
" 0.0000 \n",
" 0 \n",
" SAULXURES2 \n",
" 052465 \n",
" \n",
" \n",
" 20155 \n",
" 0.0000 \n",
" 0 \n",
" BEAUMONT-EN-VERDUNOIS \n",
" 055039 \n",
" \n",
" \n",
" 27021 \n",
" 0.0000 \n",
" 0 \n",
" ARCAY \n",
" 086008 \n",
" \n",
" \n",
" 32574 \n",
" 0.0462 \n",
" 0 \n",
" OUANARY \n",
" 102314 \n",
" \n",
" \n",
"
\n",
"
"
],
"output_type": "pyout",
"prompt_number": 182,
"text": [
" home_tax_rate home_tax_value name insee_code\n",
"35541 0.0000 0 FLEURY-DEVANT-DOUAUMONT 055189\n",
"20972 0.0331 0 MAIZERAY 055311\n",
"2701 0.0001 0 SAINT-VULBAS 001390\n",
"35998 0.0703 0 ORNES 055394\n",
"17681 0.0684 0 MOLRING 057470\n",
"3132 0.0000 0 SOULAINES-DHUYS 010372\n",
"3164 0.0000 0 VILLE-AUX-BOIS (LA) 010411\n",
"28283 0.0001 0 SAINT-COLOMBAN-DES-VILLARDS 073230\n",
"5925 0.0000 0 BOUVERANS 025085\n",
"3436 0.0001 0 CRUAS 007076\n",
"8450 0.0000 0 POMMEROL 026245\n",
"17394 0.0102 0 CHAMP-HAUT 061088\n",
"17365 0.0000 0 BOISSEI-LA-LANDE 061049\n",
"17337 0.0097 0 AUTHIEUX-DU-PUITS (LES) 061017\n",
"36596 0.0000 0 CHATEAU-D'ALMENECHES (LE ) 061101\n",
"20162 0.0000 0 BEZONVAUX 055050\n",
"10708 0.0000 0 SAULXURES2 052465\n",
"20155 0.0000 0 BEAUMONT-EN-VERDUNOIS 055039\n",
"27021 0.0000 0 ARCAY 086008\n",
"32574 0.0462 0 OUANARY 102314"
]
}
],
"prompt_number": 182
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"df[['debt_ratio']].describe()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" \n",
" debt_ratio \n",
" \n",
" \n",
" \n",
" \n",
" count \n",
" 36663.000000 \n",
" \n",
" \n",
" mean \n",
" 0.099361 \n",
" \n",
" \n",
" std \n",
" 0.134267 \n",
" \n",
" \n",
" min \n",
" 0.000000 \n",
" \n",
" \n",
" 25% \n",
" 0.039039 \n",
" \n",
" \n",
" 50% \n",
" 0.080000 \n",
" \n",
" \n",
" 75% \n",
" 0.125519 \n",
" \n",
" \n",
" max \n",
" 7.406250 \n",
" \n",
" \n",
"
\n",
"
"
],
"output_type": "pyout",
"prompt_number": 183,
"text": [
" debt_ratio\n",
"count 36663.000000\n",
"mean 0.099361\n",
"std 0.134267\n",
"min 0.000000\n",
"25% 0.039039\n",
"50% 0.080000\n",
"75% 0.125519\n",
"max 7.406250"
]
}
],
"prompt_number": 183
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# biggest debt ratio\n",
"plt.figure(figsize=(12,12));\n",
"df[['debt_ratio']].boxplot()\n",
"_df = df.sort(columns='debt_ratio', ascending=False)\n",
"_df[['debt_ratio', 'name', 'insee_code']].head(n=20)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" \n",
" debt_ratio \n",
" name \n",
" insee_code \n",
" \n",
" \n",
" \n",
" \n",
" 15510 \n",
" 7.406250 \n",
" AMBONVILLE \n",
" 052007 \n",
" \n",
" \n",
" 9472 \n",
" 4.550000 \n",
" BOUZIN \n",
" 031086 \n",
" \n",
" \n",
" 10707 \n",
" 3.718310 \n",
" SAUDRON \n",
" 052463 \n",
" \n",
" \n",
" 6236 \n",
" 3.602812 \n",
" CHAFFOIS \n",
" 025110 \n",
" \n",
" \n",
" 32363 \n",
" 3.092784 \n",
" BLANZAC-PORCHERESSE \n",
" 016046 \n",
" \n",
" \n",
" 29299 \n",
" 3.036364 \n",
" FREDIERE (LA ) \n",
" 017169 \n",
" \n",
" \n",
" 30338 \n",
" 2.989583 \n",
" BALIGNICOURT \n",
" 010027 \n",
" \n",
" \n",
" 10585 \n",
" 2.943333 \n",
" BEROU-LA-MULOTIERE \n",
" 028037 \n",
" \n",
" \n",
" 33800 \n",
" 2.904762 \n",
" PAILHAC \n",
" 065354 \n",
" \n",
" \n",
" 1913 \n",
" 2.882979 \n",
" ASSENCIERES \n",
" 010014 \n",
" \n",
" \n",
" 7157 \n",
" 2.666667 \n",
" SALEIGNES \n",
" 017416 \n",
" \n",
" \n",
" 8806 \n",
" 2.487805 \n",
" VILLEFRANCHE-LE-CHATEAU \n",
" 026375 \n",
" \n",
" \n",
" 10705 \n",
" 2.416357 \n",
" SARREY \n",
" 052461 \n",
" \n",
" \n",
" 14244 \n",
" 2.191358 \n",
" HEILTZ-L'EVEQUE \n",
" 051290 \n",
" \n",
" \n",
" 17956 \n",
" 2.162162 \n",
" BOUILH-DEVANT \n",
" 065102 \n",
" \n",
" \n",
" 7830 \n",
" 2.144330 \n",
" SANTA-LUCIA-DI-MERCURIO \n",
" 02B306 \n",
" \n",
" \n",
" 12089 \n",
" 2.044643 \n",
" DOMEYRAT \n",
" 043086 \n",
" \n",
" \n",
" 2242 \n",
" 2.009524 \n",
" EAUX-PUISEAUX \n",
" 010133 \n",
" \n",
" \n",
" 13189 \n",
" 1.993243 \n",
" SAINT-MURY-MONTEYMOND \n",
" 038430 \n",
" \n",
" \n",
" 13848 \n",
" 1.986248 \n",
" CHAUSSEE-SUR-MARNE (LA ) \n",
" 051141 \n",
" \n",
" \n",
"
\n",
"
"
],
"output_type": "pyout",
"prompt_number": 184,
"text": [
" debt_ratio name insee_code\n",
"15510 7.406250 AMBONVILLE 052007\n",
"9472 4.550000 BOUZIN 031086\n",
"10707 3.718310 SAUDRON 052463\n",
"6236 3.602812 CHAFFOIS 025110\n",
"32363 3.092784 BLANZAC-PORCHERESSE 016046\n",
"29299 3.036364 FREDIERE (LA ) 017169\n",
"30338 2.989583 BALIGNICOURT 010027\n",
"10585 2.943333 BEROU-LA-MULOTIERE 028037\n",
"33800 2.904762 PAILHAC 065354\n",
"1913 2.882979 ASSENCIERES 010014\n",
"7157 2.666667 SALEIGNES 017416\n",
"8806 2.487805 VILLEFRANCHE-LE-CHATEAU 026375\n",
"10705 2.416357 SARREY 052461\n",
"14244 2.191358 HEILTZ-L'EVEQUE 051290\n",
"17956 2.162162 BOUILH-DEVANT 065102\n",
"7830 2.144330 SANTA-LUCIA-DI-MERCURIO 02B306\n",
"12089 2.044643 DOMEYRAT 043086\n",
"2242 2.009524 EAUX-PUISEAUX 010133\n",
"13189 1.993243 SAINT-MURY-MONTEYMOND 038430\n",
"13848 1.986248 CHAUSSEE-SUR-MARNE (LA ) 051141"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAsoAAAKpCAYAAACl9gq0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nygAAUOL/A9FGNlrmk0RoAAAAAElFTkSuQmCC\n"
}
],
"prompt_number": 184
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# lowest debt ratio\n",
"_df = df.sort(columns='debt_ratio', ascending=True)\n",
"_df[['debt_ratio', 'name', 'insee_code']].head(n=20)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" \n",
" debt_ratio \n",
" name \n",
" insee_code \n",
" \n",
" \n",
" \n",
" \n",
" 6520 \n",
" 0 \n",
" SAINT-PIERRE-EN-VAUX \n",
" 021566 \n",
" \n",
" \n",
" 25748 \n",
" 0 \n",
" DONZY-LE-PERTUIS \n",
" 071181 \n",
" \n",
" \n",
" 3065 \n",
" 0 \n",
" MAS (LE ) \n",
" 006081 \n",
" \n",
" \n",
" 14705 \n",
" 0 \n",
" CELLES \n",
" 034072 \n",
" \n",
" \n",
" 14696 \n",
" 0 \n",
" CAUSSINIOJOULS \n",
" 034062 \n",
" \n",
" \n",
" 25755 \n",
" 0 \n",
" EPERTULLY \n",
" 071188 \n",
" \n",
" \n",
" 14693 \n",
" 0 \n",
" QUEUDES \n",
" 051451 \n",
" \n",
" \n",
" 25758 \n",
" 0 \n",
" ESSERTENNE \n",
" 071191 \n",
" \n",
" \n",
" 29043 \n",
" 0 \n",
" ROUY-LE-GRAND \n",
" 080683 \n",
" \n",
" \n",
" 14687 \n",
" 0 \n",
" PONTHION \n",
" 051441 \n",
" \n",
" \n",
" 14685 \n",
" 0 \n",
" POIX \n",
" 051438 \n",
" \n",
" \n",
" 14683 \n",
" 0 \n",
" POCANCY \n",
" 051435 \n",
" \n",
" \n",
" 25762 \n",
" 0 \n",
" FARGES-LES-MACON \n",
" 071195 \n",
" \n",
" \n",
" 3076 \n",
" 0 \n",
" PIERLAS \n",
" 006096 \n",
" \n",
" \n",
" 25769 \n",
" 0 \n",
" FRETTE (LA ) \n",
" 071206 \n",
" \n",
" \n",
" 9067 \n",
" 0 \n",
" BERNIENVILLE \n",
" 027057 \n",
" \n",
" \n",
" 5355 \n",
" 0 \n",
" SAVARTHES \n",
" 031537 \n",
" \n",
" \n",
" 14674 \n",
" 0 \n",
" PASSAVANT-EN-ARGONNE \n",
" 051424 \n",
" \n",
" \n",
" 14628 \n",
" 0 \n",
" MATOUGUES \n",
" 051357 \n",
" \n",
" \n",
" 14630 \n",
" 0 \n",
" MECRINGES \n",
" 051359 \n",
" \n",
" \n",
"
\n",
"
"
],
"output_type": "pyout",
"prompt_number": 185,
"text": [
" debt_ratio name insee_code\n",
"6520 0 SAINT-PIERRE-EN-VAUX 021566\n",
"25748 0 DONZY-LE-PERTUIS 071181\n",
"3065 0 MAS (LE ) 006081\n",
"14705 0 CELLES 034072\n",
"14696 0 CAUSSINIOJOULS 034062\n",
"25755 0 EPERTULLY 071188\n",
"14693 0 QUEUDES 051451\n",
"25758 0 ESSERTENNE 071191\n",
"29043 0 ROUY-LE-GRAND 080683\n",
"14687 0 PONTHION 051441\n",
"14685 0 POIX 051438\n",
"14683 0 POCANCY 051435\n",
"25762 0 FARGES-LES-MACON 071195\n",
"3076 0 PIERLAS 006096\n",
"25769 0 FRETTE (LA ) 071206\n",
"9067 0 BERNIENVILLE 027057\n",
"5355 0 SAVARTHES 031537\n",
"14674 0 PASSAVANT-EN-ARGONNE 051424\n",
"14628 0 MATOUGUES 051357\n",
"14630 0 MECRINGES 051359"
]
}
],
"prompt_number": 185
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"df[['staff_costs_ratio']].describe()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" \n",
" staff_costs_ratio \n",
" \n",
" \n",
" \n",
" \n",
" count \n",
" 36663.000000 \n",
" \n",
" \n",
" mean \n",
" 0.271105 \n",
" \n",
" \n",
" std \n",
" 0.118803 \n",
" \n",
" \n",
" min \n",
" -0.163265 \n",
" \n",
" \n",
" 25% \n",
" 0.183580 \n",
" \n",
" \n",
" 50% \n",
" 0.272727 \n",
" \n",
" \n",
" 75% \n",
" 0.354829 \n",
" \n",
" \n",
" max \n",
" 1.667286 \n",
" \n",
" \n",
"
\n",
"
"
],
"output_type": "pyout",
"prompt_number": 186,
"text": [
" staff_costs_ratio\n",
"count 36663.000000\n",
"mean 0.271105\n",
"std 0.118803\n",
"min -0.163265\n",
"25% 0.183580\n",
"50% 0.272727\n",
"75% 0.354829\n",
"max 1.667286"
]
}
],
"prompt_number": 186
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"plt.figure(figsize=(12,12));\n",
"df[['staff_costs_ratio']].boxplot()\n",
"_df = df.sort(columns='staff_costs_ratio', ascending=False)\n",
"_df[['staff_costs_ratio', 'name', 'insee_code']].head(n=20)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" \n",
" staff_costs_ratio \n",
" name \n",
" insee_code \n",
" \n",
" \n",
" \n",
" \n",
" 32562 \n",
" 1.667286 \n",
" REGINA \n",
" 102301 \n",
" \n",
" \n",
" 32571 \n",
" 0.744156 \n",
" ROURA \n",
" 102310 \n",
" \n",
" \n",
" 32597 \n",
" 0.735596 \n",
" SAINT-PHILIPPE \n",
" 104417 \n",
" \n",
" \n",
" 32575 \n",
" 0.724289 \n",
" SAUL \n",
" 102352 \n",
" \n",
" \n",
" 29228 \n",
" 0.702703 \n",
" CHAMPOULET \n",
" 045070 \n",
" \n",
" \n",
" 35052 \n",
" 0.697193 \n",
" SAINT-LOUIS \n",
" 104414 \n",
" \n",
" \n",
" 32333 \n",
" 0.695884 \n",
" FONDS-SAINT-DENIS \n",
" 103208 \n",
" \n",
" \n",
" 32567 \n",
" 0.690004 \n",
" MANA \n",
" 102306 \n",
" \n",
" \n",
" 32547 \n",
" 0.689758 \n",
" MORNE-ROUGE (LE ) \n",
" 103218 \n",
" \n",
" \n",
" 32330 \n",
" 0.672104 \n",
" CARBET (LE ) \n",
" 103204 \n",
" \n",
" \n",
" 32582 \n",
" 0.670516 \n",
" APATOU \n",
" 102360 \n",
" \n",
" \n",
" 32600 \n",
" 0.664639 \n",
" SAINTE-MARIE \n",
" 104418 \n",
" \n",
" \n",
" 32334 \n",
" 0.660211 \n",
" GRAND-RIVIERE \n",
" 103211 \n",
" \n",
" \n",
" 32569 \n",
" 0.659062 \n",
" SAINT-GEORGES \n",
" 102308 \n",
" \n",
" \n",
" 32335 \n",
" 0.658672 \n",
" FORT-DE-FRANCE \n",
" 103209 \n",
" \n",
" \n",
" 32560 \n",
" 0.657873 \n",
" MORNE-VERT (LE ) \n",
" 103233 \n",
" \n",
" \n",
" 32594 \n",
" 0.653504 \n",
" SAINT-BENOIT \n",
" 104410 \n",
" \n",
" \n",
" 36660 \n",
" 0.653467 \n",
" VAUCLIN (LE ) \n",
" 103232 \n",
" \n",
" \n",
" 32324 \n",
" 0.650403 \n",
" TERRE DE BAS \n",
" 101130 \n",
" \n",
" \n",
" 32319 \n",
" 0.649391 \n",
" POINTE NOIRE \n",
" 101121 \n",
" \n",
" \n",
"
\n",
"
"
],
"output_type": "pyout",
"prompt_number": 187,
"text": [
" staff_costs_ratio name insee_code\n",
"32562 1.667286 REGINA 102301\n",
"32571 0.744156 ROURA 102310\n",
"32597 0.735596 SAINT-PHILIPPE 104417\n",
"32575 0.724289 SAUL 102352\n",
"29228 0.702703 CHAMPOULET 045070\n",
"35052 0.697193 SAINT-LOUIS 104414\n",
"32333 0.695884 FONDS-SAINT-DENIS 103208\n",
"32567 0.690004 MANA 102306\n",
"32547 0.689758 MORNE-ROUGE (LE ) 103218\n",
"32330 0.672104 CARBET (LE ) 103204\n",
"32582 0.670516 APATOU 102360\n",
"32600 0.664639 SAINTE-MARIE 104418\n",
"32334 0.660211 GRAND-RIVIERE 103211\n",
"32569 0.659062 SAINT-GEORGES 102308\n",
"32335 0.658672 FORT-DE-FRANCE 103209\n",
"32560 0.657873 MORNE-VERT (LE ) 103233\n",
"32594 0.653504 SAINT-BENOIT 104410\n",
"36660 0.653467 VAUCLIN (LE ) 103232\n",
"32324 0.650403 TERRE DE BAS 101130\n",
"32319 0.649391 POINTE NOIRE 101121"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAtsAAAKpCAYAAABtmuBiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYRsAAEoibAMAQEn+H1/vxXa253WgAAAAAElFTkSuQmCC\n"
}
],
"prompt_number": 187
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"print df[['home_tax_rate', 'property_tax_rate', 'staff_costs_ratio', 'debt_ratio']].corr()\n",
"plot(df['property_tax_rate'], df['staff_costs_ratio'], 'o')"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
" home_tax_rate property_tax_rate staff_costs_ratio debt_ratio\n",
"home_tax_rate 1.000000 0.449842 0.179622 0.024401\n",
"property_tax_rate 0.449842 1.000000 0.400445 0.029337\n",
"staff_costs_ratio 0.179622 0.400445 1.000000 0.000091\n",
"debt_ratio 0.024401 0.029337 0.000091 1.000000\n"
]
},
{
"output_type": "pyout",
"prompt_number": 188,
"text": [
"[]"
]
},
{
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAD9CAYAAAC7iRw+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n"
}
],
"prompt_number": 188
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"df[['name', 'insee_code', 'staff_costs']].ix[df['staff_costs_ratio'].argmax()]"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 191,
"text": [
"name REGINA\n",
"insee_code 102301\n",
"staff_costs 897000\n",
"Name: 32562, dtype: object"
]
}
],
"prompt_number": 191
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### La ville de R\u00e9gina en Guyane a visiblement un ratio de charges de personnel / revenues qui est \u00e9trange (ratio de l'ordre de 1.6)"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [],
"language": "python",
"metadata": {},
"outputs": []
}
],
"metadata": {}
}
]
}