The Concept and Empirical Evidence of SWIFT Methodology
The Survey of Well-being via Instant and Frequent Tracking (SWIFT) program was created in 2014 to produce poverty statistics cost-effectively, timely, and in a user-friendly manner. Under the SWIFT program, poverty rates are estimated by (i) training a poverty rate projection model on a previous household budget survey, (ii) collecting data in the field on identified poverty correlates, and (iii) applying the model to collected data to produce poverty projections. The SWIFT program has expanded since 2014 and now includes a wide variety of activities, namely, aiding in estimating official poverty statistics and monitoring poverty outcomes of lending operations and investment projects in over 50 countries. One key advantage of the SWIFT methodology is the significantly shorter interview time compared to the surveys used in traditional poverty data collection. In 2015, the SWIFT program prepared a guideline for using the SWIFT methodology to ensure the quality and precision of poverty estimation under the program. The guideline includes a set of recommendations on data collection and poverty projection. This report documents the performance evaluation and possible modifications of the SWIFT poverty projection methods since 2015. It critically evaluates the SWIFT poverty projection method proposed by the 2015 guideline, identifies limitations, and introduces several remedies – SWIFT Plus and SWIFT 2.0 – with their evaluations. This paper is organized as follows: Section 2 reviews the literature, Section 3 describes the SWIFT methodology, Section 4 discusses challenges for SWIFT and the solutions, Section 5 explores future research topics, and Section 6 concludes.
Main Author: | |
---|---|
Format: | Working Paper biblioteca |
Language: | English en_US |
Published: |
Washington, DC
2022-06
|
Subjects: | SWIFT, CROSS-VALIDATION (CV), MULTIPLE IMPUTATION (MI), POVERTY RATES, SIMULATION AND ESTIMATION, MODEL STABILITY, RAPID CONSUMPTION SURVEY, ELL METHOD, PREDICTIVE MEAN MATCHING (PMM), MACHINE LEARNING, BAYESIAN FRAMEWORK, |
Online Access: | http://documents.worldbank.org/curated/en/099547109302235758/IDU04a3a086c0a2da04853084b10e855253105f9 http://hdl.handle.net/10986/38095 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
dig-okr-1098638095 |
---|---|
record_format |
koha |
spelling |
dig-okr-10986380952022-10-03T05:10:43Z The Concept and Empirical Evidence of SWIFT Methodology World Bank SWIFT CROSS-VALIDATION (CV) MULTIPLE IMPUTATION (MI) POVERTY RATES SIMULATION AND ESTIMATION MODEL STABILITY RAPID CONSUMPTION SURVEY ELL METHOD PREDICTIVE MEAN MATCHING (PMM) MACHINE LEARNING BAYESIAN FRAMEWORK The Survey of Well-being via Instant and Frequent Tracking (SWIFT) program was created in 2014 to produce poverty statistics cost-effectively, timely, and in a user-friendly manner. Under the SWIFT program, poverty rates are estimated by (i) training a poverty rate projection model on a previous household budget survey, (ii) collecting data in the field on identified poverty correlates, and (iii) applying the model to collected data to produce poverty projections. The SWIFT program has expanded since 2014 and now includes a wide variety of activities, namely, aiding in estimating official poverty statistics and monitoring poverty outcomes of lending operations and investment projects in over 50 countries. One key advantage of the SWIFT methodology is the significantly shorter interview time compared to the surveys used in traditional poverty data collection. In 2015, the SWIFT program prepared a guideline for using the SWIFT methodology to ensure the quality and precision of poverty estimation under the program. The guideline includes a set of recommendations on data collection and poverty projection. This report documents the performance evaluation and possible modifications of the SWIFT poverty projection methods since 2015. It critically evaluates the SWIFT poverty projection method proposed by the 2015 guideline, identifies limitations, and introduces several remedies – SWIFT Plus and SWIFT 2.0 – with their evaluations. This paper is organized as follows: Section 2 reviews the literature, Section 3 describes the SWIFT methodology, Section 4 discusses challenges for SWIFT and the solutions, Section 5 explores future research topics, and Section 6 concludes. 2022-10-02T15:03:24Z 2022-10-02T15:03:24Z 2022-06 Working Paper http://documents.worldbank.org/curated/en/099547109302235758/IDU04a3a086c0a2da04853084b10e855253105f9 http://hdl.handle.net/10986/38095 English en_US Equitable Growth, Finance and Institutions Insight - Poverty and Equity; CC BY 3.0 IGO http://creativecommons.org/licenses/by/3.0/igo World Bank Washington, DC Working Papers Working Papers :: Other Papers World |
institution |
Banco Mundial |
collection |
DSpace |
country |
Estados Unidos |
countrycode |
US |
component |
Bibliográfico |
access |
En linea |
databasecode |
dig-okr |
tag |
biblioteca |
region |
America del Norte |
libraryname |
Biblioteca del Banco Mundial |
language |
English en_US |
topic |
SWIFT CROSS-VALIDATION (CV) MULTIPLE IMPUTATION (MI) POVERTY RATES SIMULATION AND ESTIMATION MODEL STABILITY RAPID CONSUMPTION SURVEY ELL METHOD PREDICTIVE MEAN MATCHING (PMM) MACHINE LEARNING BAYESIAN FRAMEWORK SWIFT CROSS-VALIDATION (CV) MULTIPLE IMPUTATION (MI) POVERTY RATES SIMULATION AND ESTIMATION MODEL STABILITY RAPID CONSUMPTION SURVEY ELL METHOD PREDICTIVE MEAN MATCHING (PMM) MACHINE LEARNING BAYESIAN FRAMEWORK |
spellingShingle |
SWIFT CROSS-VALIDATION (CV) MULTIPLE IMPUTATION (MI) POVERTY RATES SIMULATION AND ESTIMATION MODEL STABILITY RAPID CONSUMPTION SURVEY ELL METHOD PREDICTIVE MEAN MATCHING (PMM) MACHINE LEARNING BAYESIAN FRAMEWORK SWIFT CROSS-VALIDATION (CV) MULTIPLE IMPUTATION (MI) POVERTY RATES SIMULATION AND ESTIMATION MODEL STABILITY RAPID CONSUMPTION SURVEY ELL METHOD PREDICTIVE MEAN MATCHING (PMM) MACHINE LEARNING BAYESIAN FRAMEWORK World Bank The Concept and Empirical Evidence of SWIFT Methodology |
description |
The Survey of Well-being via Instant
and Frequent Tracking (SWIFT) program was created in 2014 to
produce poverty statistics cost-effectively, timely, and in
a user-friendly manner. Under the SWIFT program, poverty
rates are estimated by (i) training a poverty rate
projection model on a previous household budget survey, (ii)
collecting data in the field on identified poverty
correlates, and (iii) applying the model to collected data
to produce poverty projections. The SWIFT program has
expanded since 2014 and now includes a wide variety of
activities, namely, aiding in estimating official poverty
statistics and monitoring poverty outcomes of lending
operations and investment projects in over 50 countries. One
key advantage of the SWIFT methodology is the significantly
shorter interview time compared to the surveys used in
traditional poverty data collection. In 2015, the SWIFT
program prepared a guideline for using the SWIFT methodology
to ensure the quality and precision of poverty estimation
under the program. The guideline includes a set of
recommendations on data collection and poverty projection.
This report documents the performance evaluation and
possible modifications of the SWIFT poverty projection
methods since 2015. It critically evaluates the SWIFT
poverty projection method proposed by the 2015 guideline,
identifies limitations, and introduces several remedies –
SWIFT Plus and SWIFT 2.0 – with their evaluations. This
paper is organized as follows: Section 2 reviews the
literature, Section 3 describes the SWIFT methodology,
Section 4 discusses challenges for SWIFT and the solutions,
Section 5 explores future research topics, and Section 6 concludes. |
format |
Working Paper |
topic_facet |
SWIFT CROSS-VALIDATION (CV) MULTIPLE IMPUTATION (MI) POVERTY RATES SIMULATION AND ESTIMATION MODEL STABILITY RAPID CONSUMPTION SURVEY ELL METHOD PREDICTIVE MEAN MATCHING (PMM) MACHINE LEARNING BAYESIAN FRAMEWORK |
author |
World Bank |
author_facet |
World Bank |
author_sort |
World Bank |
title |
The Concept and Empirical Evidence of SWIFT Methodology |
title_short |
The Concept and Empirical Evidence of SWIFT Methodology |
title_full |
The Concept and Empirical Evidence of SWIFT Methodology |
title_fullStr |
The Concept and Empirical Evidence of SWIFT Methodology |
title_full_unstemmed |
The Concept and Empirical Evidence of SWIFT Methodology |
title_sort |
concept and empirical evidence of swift methodology |
publisher |
Washington, DC |
publishDate |
2022-06 |
url |
http://documents.worldbank.org/curated/en/099547109302235758/IDU04a3a086c0a2da04853084b10e855253105f9 http://hdl.handle.net/10986/38095 |
work_keys_str_mv |
AT worldbank theconceptandempiricalevidenceofswiftmethodology AT worldbank conceptandempiricalevidenceofswiftmethodology |
_version_ |
1756576194481881088 |