povertyInequalityMeasures 1.1.0

Calculate some poverty and inequality measures. Based on the Wold Bank Poverty and Inequality Handbook: https://documents1.worldbank.org/curated/en/488081468157174849/pdf/483380PUB0Pove101OFFICIAL0USE0ONLY1.pdf

Poverty and Inequality Measures

Python Package to calculate some poverty and inequality measures.

Based on the Wold Bank Poverty and Inequality Handbook.

Poverty

Package includes calculations of:

• Headcount Index: get_headcount_index (pg 68 of the Handbook)
• Poverty Gap Index: get_poverty_gap_index (pg 70)
• Poverty Severity Index: get_poverty_severity_index (pg 71)
• Generic Poverty Severity Index: get_poverty_severity_index_generic (pg 72)
• Sen Index: get_sen_index (pg 74)
• Watts Index: get_watts_index (pg 77)
• Time To Exit Poverty: get_time_to_exit (pg 78)

Inequality

Package includes calculations of:

• Gini Coefficient: get_gini (pg 104)
• Palma Ratio: get_palma (not in the handbook but defined as being the ratio between the income or expenditure of the richest decile divided by the income or expenditure of the poorest four deciles)

Installation

pip install povertyInequalityMeasures

General Usage

All methods require a data dataframe with at least two columns:

1. target_col: The column of data that is being used to measure poverty/inequality. In most cases this is either some sort of total household/individual expenditure or income. See page 20 of the handbook for considerations of which to use.
2. weight_col: The column that represents the weighting of each row of data. Normally, data used is survey data and therefore each row (a household or individual sureveyed) represents a given number of actual households/individuals in the population. The weight column should hold that information.

Additionally all poverty methods require a poverty line (pl) parameter, which is the amount of expenditure/income below which someone is said to be in poverty.

The time to exit method requires a growth parameter for the expected growth rate of the economy over time.

The Generic Poverty Severity Index method requires an alpha parameter, which must be greater than 0.

Examples

from povertyInequalityMeasures import poverty, inequality
import pandas as pd

Example 1

data = pd.DataFrame({'total_expenditure': [ 100,110,150,160], "weight":[1,1,1,1]})

result=poverty.get_headcount_index(125,data,"total_expenditure","weight")

print(result)
#0.5

Example 2

data = pd.DataFrame({'total_expenditure': [ 100,110,150,160], 'weight':[1,1,1,1]})
poverty_line= 125
result = poverty.get_sen_index(poverty_line, data, "total_expenditure","weight")


print(result)
#0.374

Example 3

data = pd.DataFrame({'total_expenditure': [ 100,110,150,160], 'weight':[1,1,1,1]})
poverty_line= 125
result = poverty.get_poverty_severity_index_generic(poverty_line, data, "total_expenditure","weight",2)


print(result)
#0.0136

Example 4

data = pd.DataFrame({'total_expenditure': [7,10,15,18], 'weight':[1,1,1,1]})

result = inequality.get_gini(data, "total_expenditure","weight")

print(result)
#0.19