Exemplo n.º 1
0
 def __init__(self):
     self.query = queries(
     )  #query object has query texts for all database interactions as different functions
     self.parse = parse_inputs(
     )  #parses the LAR row into selected components and stores them in a dictionary
     self.connect = connect_DB(
     )  #sets a database connection and passes a cursor back
     self.orig_by_tract_minority_category = [
     ]  #sorted alphabetically: high, low, middle, upper
     self.app_by_tract_minority_category = [
     ]  #sorted alphabetically: high, low, middle, upper
     self.tract_orig_rates = [
     ]  #holds an alphabetically sorted list of origination rates for tract categorized by minority population
     #order of tract_orig_rates list: high, low, middle, upper
     self.nonminority_approval_rate = 0  #establishes a variable to hold rates for graphing
     self.HMDA_cols = [
         'statecode', 'countycode', 'censustractnumber', 'applicantrace1',
         'applicantrace2', 'applicantrace3', 'applicantrace4',
         'applicantrace5', 'coapplicantrace1', 'coapplicantrace2',
         'coapplicantrace3', 'coapplicantrace4', 'coapplicantrace5',
         'applicantethnicity', 'coapplicantethnicity', 'applicantincome',
         'ratespread', 'lienstatus', 'hoepastatus', 'purchasertype',
         'loanamount', 'asofdate', 'hud_median_family_income',
         'minority_population_pct', 'tract_to_msa_md_income'
     ]
Exemplo n.º 2
0
from pylab import *
from sklearn.cross_validation import cross_val_score
#from years_of_milwaukee import call_2009
#from years_of_milwaukee import call_2010
#from years_of_milwaukee import call_2011
#from years_of_milwaukee import call_2012
#from years_of_milwaukee import call_2013
#from years_of_milwaukee import call_tracts_2013 as tracts_2013
#from years_of_milwaukee import call_tracts_2012 as tracts_2012
#from years_of_milwaukee import call_tracts_2011 as tracts_2011
#from years_of_milwaukee import call_tracts_2010 as tracts_2010
#from years_of_milwaukee import call_tracts_2009 as tracts_2009
#instantiate classes
#geo = geo_aggregator() #disabled geo_aggregator as it has been used and is not needed
query = queries()
connect = connect_DB()
parse = parse_inputs()

HMDA_cols = ['statecode', 'countycode', 'censustractnumber', 'applicantrace1', 'applicantrace2', 'applicantrace3', 'applicantrace4', 'applicantrace5',
                            'coapplicantrace1', 'coapplicantrace2', 'coapplicantrace3', 'coapplicantrace4', 'coapplicantrace5', 'applicantethnicity', 'coapplicantethnicity',
                            'applicantincome', 'ratespread', 'lienstatus', 'hoepastatus', 'purchasertype', 'loanamount', 'year', 'median_family_income',
                            'minority_population_pct', 'tract_to_msa_md_income', 'actiontype', 'gender']

'''data have been filtered as follows in SQL queries
    loan type: conventional only, code 1
    property type: 1-4 family only, code 1
    loan purpose: purchase only, code 1
    occupancy status: owner occupied only, code 1
    lien status: first liens only, code 1
    income != 0 #removes most non-natural person loans
    action type: originations includes only originated loans, applications includes codes 1-4
Exemplo n.º 3
0
#from geo_aggregator_2013 import geo_aggregator
from DAT4_library import connect_DB
from DAT4_library import parse_inputs
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from pylab import *
from years_of_milwaukee import call_2009
from years_of_milwaukee import call_2010
from years_of_milwaukee import call_2011
from years_of_milwaukee import call_2012
from years_of_milwaukee import call_2013
#instantiate classes
#geo = geo_aggregator() #disabled geo_aggregator as it has been used and is not needed
query = queries()
connect = connect_DB()
parse = parse_inputs()

#mil_2009 = call_2009()
#mil_2010 = call_2010()
#mil_2011 = call_2011()
#mil_2012 = call_2012()
mil_2013 = call_2013()


class year_calls(object):
    pass


class call_tracts_2013(year_calls):
    def __init__(self):