hbsir 0.4.4

An intuitive platform designed to simplify the analysis of Iran household budget survey data.

HBSIR - Household Budget Survey of Iran in Python

Overview

HBSIR is a Python package for working with microdata from the Household Budget Survey of Iran (HBSIR). It provides a high-level API for easily loading, exploring and analyzing this nationally representative household survey.

The key capabilities of HBSIR include:

• Loading survey tables for multiple years in raw, cleaned or normalized form
• Enriching data by adding geographical attributes and classification systems
• Performing correct weighted statistical analysis with integrated sampling weights
• Calculating metrics like weighted averages and distributional statistics
• Accessing pre-processed tables and visualizations stored in a Knowledge Base
• Handling all necessary data processing and configuration behind the scenes

HBSIR aims to make analysis of the Household Budget Survey straightforward for researchers, analysts and policymakers even without programming expertise. The simple verb-based API provides a smooth on-ramp while harnessing the full power of Pandas and Python under the hood.

Motivation

The HBSIR Data Python Package addresses the historical challenges in analyzing Household Budget Survey data from Iran. The complex nature of the data, with its ever-changing codings, inconsistent table schemas, and lack of standardization, has made analysis a daunting task. This package aims to simplify the entire process—from data extraction to analysis—allowing researchers to focus on deriving meaningful insights rather than wrestling with data complexities.

Recognizing the difficulties posed by the original MS Access format and the lack of standardization across different years, the HBSIR Data Python Package offers a solution. By providing a uniform API, streamlining data loading, and introducing a customizable schema system, we aim to make HBSIR data more accessible and usable for all researchers. This package empowers users to overcome the challenges associated with HBSIR data, enabling them to contribute to a deeper understanding of household budget dynamics in Iran.

Getting Started

Getting started with HBSIR is quick and easy.

First, install the package using pip:

pip install hbsir

Then import the library and load a table:

import hbsir

table = hbsir.load_table("Expenditures", years=[1390, 1400])

The load_table function, by default, retrieves normalized, analysis-ready data. Just specify the table name and year(s).

Further data enrichment occurs through HBSIR's verb-based API:

table = hbsir.add_attribute(table, name="Province")

This adds a province name column for geographical analysis.

For more examples of data loading, transformation and analysis refer to the complete documentation.

Contributing

HBSIR warmly welcomes contributions from the community!

As an open source project aimed at accessibility, we highly encourage submissions and improvements from all users.

In particular, assistance with updating and expanding the documentation is invaluable. As the capabilities of HBSIR grow, keeping the docs updated is an ongoing effort. All contributions to the documentation are appreciated.

More broadly, please open issues or pull requests for:

• Reporting bugs
• Suggesting new features
• Improving existing code

The HBSIR contributors will do their best to respond promptly and incorporate changes that align with the project's overall vision and aims.

As an open dataset and software project, transparency and community participation are vital to the success and impact of HBSIR. We welcome you to get involved!

Related Packages

HBSIR is built on top of the BSSIR (Basic Survey Structure of Iran) library which provides shared data loading and transformation functionality across multiple socioeconomic surveys from Iran.

Specifically, HBSIR utilizes capabilities from BSSIR to:

• Read complex raw survey data files
• Standardize schemas across years
• Integrate metadata like regional attributes
• Manage data configurations

This enables HBSIR to focus specifically on provisions for analyzing the Household Budget Survey without having to reimplement common data handling mechanisms.

The modular design centered around BSSIR also allows the HBSIR data access API to be reused easily. Two other survey-specific libraries that inherit directly from HBSIR are:

• LFSIR: For the Iran Labor Force Survey
• CNSIR: For the Iran Census and Sample Survey

So in summary, BSSIR provides base survey data infrastructure, while HBSIR, LFSIR and CNSIR build on top of it with added provisions for analyzing their respective datasets. This hierarchy of packages allows collaborative development and maintenance.

flowchart TD
    BSSIR --> HBSIR
    BSSIR --> LFSIR
    BSSIR --> CNSIR

    click BSSIR "https://github.com/Iran-Open-Data/BSSIR"
    click HBSIR "https://github.com/Iran-Open-Data/HBSIR"
    click LFSIR "https://github.com/Iran-Open-Data/LFSIR"
    click CNSIR "https://github.com/Iran-Open-Data/CNSIR"