def getWBDataFromWeb(self, pStockCode, pStart, pEnd):
#https://wbdata.readthedocs.io/en/latest/
wbdata.get_source()
wbdata.get_indicator(source=1)
wbdata.search_countries("united")
date = (datetime.datetime(2010, 1, 1), datetime.datetime(2011, 1, 1))
self.data = wbdata.get_data("IC.BUS.EASE.XQ",
country=("USA", "GBR"),
data_date=date)
for row in self.data:
print(row['country']['id'], row)
#indicators = {"IC.BUS.EASE.XQ": "doing_business", "NY.GDP.PCAP.PP.KD": "gdppc"}
#df = wbdata.get_dataframe(indicators, country=countries, convert_date=True)
#df.describe()
return self.data
def testOneTopic(self):
wbdata.get_source("3")
def testGetAllTopics(self):
wbdata.get_source()
def testOneSource(self):
wbdata.get_source(31)
def testGetAllSources(self):
wbdata.get_source()
def plotNANperindicator(data, background=False):
# PLOT NUMBER OF NAN PER INDICATOR
indicators = list(set(data.columns.levels[1]))
# Checks how many nans per indicator
how_many_nans = []
for i in indicators:
temp = data.xs(i, level=1, axis=1).isnull().astype(int).values.sum()
how_many_nans.append(temp)
# Sorts the indicators from most nan to less nan
ind = np.argsort(how_many_nans)[::-1]
if (background):
#create labels2 dictt
sources = wbdata.get_source(display=False)
id_to_sourceName = dict(
zip([k['id'] for k in sources], [k['name'] for k in sources]))
all_indics = wbdata.get_indicator(display=False)
indicator_to_id = dict(
zip([k['id'] for k in all_indics],
[k['source']['id'] for k in all_indics]))
for i in (set(indicators) - set(indicator_to_id.keys())):
indicator_to_id[i] = -1
id_to_sourceName[-1] = 'NaN'
# Gets labels, sorts them, puts them in (int) form
range_indics = [indicators[k]
for k in ind] # sorted indicator code (by most nans):
range_labels_int = [indicator_to_id[key] for key in range_indics]
range_labels = [
id_to_sourceName[indicator_to_id[key]] for key in range_indics
]
# Define cmap for coloring the labels
cmap = plt.get_cmap('jet_r')
color = cmap(np.linspace(0, 1.0, len(set(range_labels_int))))
# Helper function to get arrays of consecutive values, to plot colors on xaxis, used below
def consecutive(data, stepsize=1):
return np.split(data, np.where(np.diff(data) != stepsize)[0] + 1)
# Line plot of number of nan values per country, with background colored according to indicator label
plt.figure(figsize=(10, 5))
plt.plot(range(0, len(indicators)), [
100 * (how_many_nans[k] / data.xs('SP.POP.TOTL', level=1, axis=1).size)
for k in ind
])
if (background):
plt.title(
'Number of NaN values per indicator, sorted (background = indicator label)',
fontsize=15)
else:
plt.title('Number of NaN values per indicator, sorted', fontsize=15)
plt.xlabel('Indicators', fontsize=15)
plt.ylabel('# NaN values (% of total)', fontsize=15)
plt.xticks([])
if (background):
# Prepare patches (to color the background according to the indicator label), and the legend
legends = []
a = 0
for i in list(
set(range_labels_int)): # Let's say we have 4 labels : i=0:3
index_country_label = [
k for k in range(0, len(range_labels_int))
if range_labels_int[k] == i
] # we get the index of each indicator with label i
index_country_label = consecutive(
index_country_label
) # we get the consecutive indexes. For instance consecutive([1,2,3,5,7,8,9]) = [[1,3],[5],[7,9]]
patch = mpatches.Patch(
color=color[a], alpha=0.3
) # Colors the background of each indicator according to its label
legends.append(patch)
for j in range(0, len(index_country_label)):
temp = len(index_country_label[j])
plt.axvspan(index_country_label[j][0],
index_country_label[j][temp - 1],
color=color[a],
alpha=0.3,
lw=2.0)
a += 1
plt.legend(handles=legends,
loc='upper center',
bbox_to_anchor=(0.5, -0.06),
fancybox=True,
shadow=True,
ncol=int(
(1 / 5) * len(np.unique(range_labels_int))),
fontsize='small')
plt.show()
def testOneSource(self):
wbdata.get_source(31)
def get_categories():
categories = [(x['id'], x['name']) for x in wbdata.get_source(display=False)]
return categories
# Keep_levels
# if True don’t reduce the number of index levels returned if only getting one date or country
# Cache
# use the cache
# Returns
# a WBDataFrame
import pandas as pd
import wbdata as wb
import datetime
# search for data sources in world bank data
wb.get_source()
wb.get_indicator(source=16)
# do country search
wb.search_countries('united')
# do wild search
wb.search_countries('niger*')
# get data for country
# SE.ADT.1524.LT.FM.ZS Literacy rate, youth (ages 15-24), gender parity index (GPI)
# return a multi-dictionary(based on year) list
wb.get_data("SE.ADT.1524.LT.FM.ZS", country="USA")
# selecting data range
date_range = datetime.datetime(2008, 1, 1), datetime.datetime(2019, 1, 1)
def testGetAllSources(self):
wbdata.get_source()
# Created to get information from World Bank API
import wbdata
# All available topics
available_info = wbdata.get_source()
for x in available_info:
print(x['id'], x['name'])
# Information about G20 Financial Inclusion Indicators
inclusion_indicators = wbdata.get_indicator(source=33)
print(inclusion_indicators)
# Information 'The consumer price index reflects the
# change in prices for the average consumer of a constant basket of consumer
# goods. Data is in nominal terms and seasonally adjusted.' for USA for all
# years.
cpi = wbdata.get_data('CPTOTSAXN', country='USA')
print(cpi)
import wbdata as wb results = wb.get_source() wb.get_indicator(source=1)
def search_sources(self):
wb.get_source()
import wbdata from pprint import pprint news = wbdata.get_source() pprint(news)
def testGetAllTopics(self):
wbdata.get_source()
# Gives data in reverse order by default
data1 = wbd.get_dataframe(indicator1, country1).sort_index()
data1.head()
data1.plot()
# This is fine but what if you need to find different countries?
wbd.get_country()
# Too long a list, easier to search
wbd.search_countries('South')
# What if you want to get different indicators
#wbd.get_indicator() # Too slow
wbd.search_indicators('GDP') # Too many!
# Perhaps instead look by source
wbd.get_source()
# or topic
wbd.get_topic()
# Now search
wbd.search_indicators('CO2', topic=19)
# What about getting multiple countries
country2 = ['IE', 'US', 'CN'] # Ireland, USA, China
indicator2 = {'EN.ATM.CO2E.KT': 'CO2 emissions (kt)'}
# Get the data
data2 = wbd.get_dataframe(indicator2, country2).sort_index()
# Need to unstack to get this into proper order
data2_u = data2.unstack(level=0)
data2_u.head()
data2_u.plot()
def testOneTopic(self):
wbdata.get_source("3")
#ease of business
import wbdata
wbdata.get_source()
wbdata.get_indicator(source=1)
#get country codes with a search
wbdata.search_countries('Turkey') #TUR
wbdata.get_data('IC.REG.COST.PC.MA.ZS', country='TUR')[0]
wbdata.search_countries('united') #GBR
wbdata.get_data('IC.REG.COST.PC.MA.ZS', country='GBR')
import datetime
data_date = (datetime.datetime(2010, 1, 1), datetime.datetime(2011, 1, 1))
wbdata.get_data("IC.REG.COST.PC.MA.ZS",
country=("USA", "GBR"),
data_date=data_date)
wbdata.search_indicators("gdp per capita")
wbdata.get_data('NY.GDP.PCAP.KD.ZG')
wbdata.get_data('NY.GDP.PCAP.KD.ZG', country='USA')
wbdata.get_data('NY.GDP.PCAP.KD.ZG', country='OED')
#income level filter
wbdata.get_incomelevel()
countries = [
i['id'] for i in wbdata.get_country(incomelevel="HIC", display=False)
]
indicators = {
"IC.REG.COST.PC.MA.ZS": "doing_business",
"NY.GDP.PCAP.PP.KD": "gdppc"
}
import wbdata
import datetime as dt
indicatorSelection = {"DT.DOD.DLXF.CD": "ExternalDebtStock"}
# NY.GDP.DEFL.KD.ZG
locationSelection = ["US", "SSA", "SAS", "LAC", "MNA", "EAP"]
timeSelection = (dt.datetime(2009, 1, 1), dt.datetime(2018, 12, 31))
IDS = wbdata.get_source(source_id=6)
print(IDS)
#grab indicators above for countires above and load into data frame
df = wbdata.get_dataframe(indicators, country=countries, convert_date=False)
#df is "pivoted", pandas' unstack fucntion helps reshape it into something plottable
dfu = df.unstack(level=0)
# a simple matplotlib plot with legend, labels and a title
dfu.plot();
plt.legend(loc='best');
plt.title("GNI Per Capita ($USD, Atlas Method)");
plt.xlabel('Date'); plt.ylabel('GNI Per Capita ($USD, Atlas Method');
# In[ ]:
wbdata.get_source()
# In[ ]:
# 1 DOing Business
wbdata.get_indicator(source=1)
# In[ ]:
wbdata.search_countries("Brazil")
# In[ ]:
def print_wb_sources():
wbdata.get_source()