It Can’t Get No Worse: Using Twitter Data to Improve GDP Estimates for Developing Countries

<p dir="ltr">This paper shows that we can use social media data to improve the accuracy of GDP estimates at the country level for developing countries. I use all publicly available image tweets from 2012 and 2013 to estimate GDP at the country level for developing countries. First, I...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Agustín Indaco (21633494) (author)
منشور في: 2023
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513522140971008
author Agustín Indaco (21633494)
author_facet Agustín Indaco (21633494)
author_role author
dc.creator.none.fl_str_mv Agustín Indaco (21633494)
dc.date.none.fl_str_mv 2023-07-04T03:00:00Z
dc.identifier.none.fl_str_mv 10.3390/engproc2023039049
dc.relation.none.fl_str_mv https://figshare.com/articles/conference_contribution/It_Can_t_Get_No_Worse_Using_Twitter_Data_to_Improve_GDP_Estimates_for_Developing_Countries/31444201
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Economics
Applied economics
Information and computing sciences
Information systems
Mathematical sciences
Statistics
national accounts
social media data
nowcasting
dc.title.none.fl_str_mv It Can’t Get No Worse: Using Twitter Data to Improve GDP Estimates for Developing Countries
dc.type.none.fl_str_mv Text
Conference contribution
info:eu-repo/semantics/publishedVersion
text
conference object
description <p dir="ltr">This paper shows that we can use social media data to improve the accuracy of GDP estimates at the country level for developing countries. I use all publicly available image tweets from 2012 and 2013 to estimate GDP at the country level for developing countries. First, I find that one can explain 76% of the cross-country variation in GDP with the volume of tweets sent from each country. I then show that the residuals on these Twitter-GDP estimates are significantly larger for countries with allegedly poor data quality. I then use Nigeria as a case study to show that this method delivers much more timely and accurate estimates than those presented by official statistic agencies.</p><h2 dir="ltr">Other Information</h2><p dir="ltr">Published in: ITISE 2023<br>License: <a href="https://creativecommons.org/licenses/by/4.0/" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.3390/engproc2023039049" target="_blank">https://dx.doi.org/10.3390/engproc2023039049</a></p>
eu_rights_str_mv openAccess
id Manara2_0a057db6b49b0d34ab02659d2f084cc5
identifier_str_mv 10.3390/engproc2023039049
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/31444201
publishDate 2023
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling It Can’t Get No Worse: Using Twitter Data to Improve GDP Estimates for Developing CountriesAgustín Indaco (21633494)EconomicsApplied economicsInformation and computing sciencesInformation systemsMathematical sciencesStatisticsnational accountssocial media datanowcasting<p dir="ltr">This paper shows that we can use social media data to improve the accuracy of GDP estimates at the country level for developing countries. I use all publicly available image tweets from 2012 and 2013 to estimate GDP at the country level for developing countries. First, I find that one can explain 76% of the cross-country variation in GDP with the volume of tweets sent from each country. I then show that the residuals on these Twitter-GDP estimates are significantly larger for countries with allegedly poor data quality. I then use Nigeria as a case study to show that this method delivers much more timely and accurate estimates than those presented by official statistic agencies.</p><h2 dir="ltr">Other Information</h2><p dir="ltr">Published in: ITISE 2023<br>License: <a href="https://creativecommons.org/licenses/by/4.0/" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.3390/engproc2023039049" target="_blank">https://dx.doi.org/10.3390/engproc2023039049</a></p>2023-07-04T03:00:00ZTextConference contributioninfo:eu-repo/semantics/publishedVersiontextconference object10.3390/engproc2023039049https://figshare.com/articles/conference_contribution/It_Can_t_Get_No_Worse_Using_Twitter_Data_to_Improve_GDP_Estimates_for_Developing_Countries/31444201CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/314442012023-07-04T03:00:00Z
spellingShingle It Can’t Get No Worse: Using Twitter Data to Improve GDP Estimates for Developing Countries
Agustín Indaco (21633494)
Economics
Applied economics
Information and computing sciences
Information systems
Mathematical sciences
Statistics
national accounts
social media data
nowcasting
status_str publishedVersion
title It Can’t Get No Worse: Using Twitter Data to Improve GDP Estimates for Developing Countries
title_full It Can’t Get No Worse: Using Twitter Data to Improve GDP Estimates for Developing Countries
title_fullStr It Can’t Get No Worse: Using Twitter Data to Improve GDP Estimates for Developing Countries
title_full_unstemmed It Can’t Get No Worse: Using Twitter Data to Improve GDP Estimates for Developing Countries
title_short It Can’t Get No Worse: Using Twitter Data to Improve GDP Estimates for Developing Countries
title_sort It Can’t Get No Worse: Using Twitter Data to Improve GDP Estimates for Developing Countries
topic Economics
Applied economics
Information and computing sciences
Information systems
Mathematical sciences
Statistics
national accounts
social media data
nowcasting