The methodologies adopted by national statistical offices to collect data of within- and cross-country Input-Output (I/O) economic relations raises the issue of obtaining reliable data in a timely fashion, making the reconstruction of I/O tables of particular interest. In this work, we propose a method combining Hierarchical Clustering and Matrix Completion to impute missing entries of a partially unknown I/O matrix. Through a validation study based on World Input-Output Database as well as on synthetic matrices, we show the effectiveness of the proposed method to impute missing values from both previous years data and current data related to countries similar to the one for which current data are missing.
Le metodologie adottate dagli uffici statistici nazionali per raccogliere i dati sulle relazioni economiche di Input-Output (I/O) pongono il problema di ot- tenere dati affidabili in modo tempestivo. La ricostruzione delle tabelle I/O risulta perci`o di particolare interesse. Proponiamo un approccio statistico che combina clustering gerarchica e Matrix Completion allo scopo di imputare i dati mancanti di una matrice I/O. Attraverso uno studio di validazione basato su World Input- Output Database e matrici sintetiche, mostriamo l’efficacia del metodo proposto per imputare i valori mancanti utilizzando sia dati dello stesso paese negli anni precedenti sia dati dello stesso periodo relativi a paesi simili a quello che presenta dati mancanti.
(2022). A Statistical Approach for the Completion of Input-Output Tables . Retrieved from http://hdl.handle.net/10446/229930
A Statistical Approach for the Completion of Input-Output Tables
Metulini, Rodolfo;
2022-01-01
Abstract
The methodologies adopted by national statistical offices to collect data of within- and cross-country Input-Output (I/O) economic relations raises the issue of obtaining reliable data in a timely fashion, making the reconstruction of I/O tables of particular interest. In this work, we propose a method combining Hierarchical Clustering and Matrix Completion to impute missing entries of a partially unknown I/O matrix. Through a validation study based on World Input-Output Database as well as on synthetic matrices, we show the effectiveness of the proposed method to impute missing values from both previous years data and current data related to countries similar to the one for which current data are missing.File | Dimensione del file | Formato | |
---|---|---|---|
Metulini_et_al_2022_SIS_shortversion.pdf
Solo gestori di archivio
Versione:
publisher's version - versione editoriale
Licenza:
Licenza default Aisberg
Dimensione del file
791.73 kB
Formato
Adobe PDF
|
791.73 kB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
Aisberg ©2008 Servizi bibliotecari, Università degli studi di Bergamo | Terms of use/Condizioni di utilizzo