Disaggregating input–output tables by the multidimensional RAS method: a case study of the Czech Republic

Eficiencia en la estimación de coeficientes técnicos y multiplicadores interregionales: la metodología Jahn versus las metodologías GRAS y GravityRAS

“…On the other hand, bi-proportional techniques such as the RAS (Stone, 1961) or some of its variants are widely used, such as the Generalized-RAS or GRAS, whose advantage lies in being implementable when the table contains both positive and negative values (Günlük-Senesen &Bates, 1988 andJunius &Oosterhaven, 2003); the Cell-corrected RAS (Mínguez et al, 2009) , which uses cell variation distributions computed from multiple matrices of different periods or different regions, to modify the RAS solution by solving an additional optimisation problem that produces the most likely cell corrections; or PATH-RAS (Pereira-López et al, 2013), that can be applied to rectangular matrices and has minimal information requirements. Likewise, several works have provided improvements to the GRAS methodology: correcting the objective function (Huang et al, 2008;Lemelin, 2009;Lenzen et al, 2007); ensuring the fulfilment of some constraints infeasible by other RAS methods, through an iterative method that allows changing the sign in successive iterations (Lenzen, Moran, et al, 2014;Temurshoev et al, 2013); working with multidimensional tables (Valderas-Jaramillo & Rueda-Cantuche, 2021;Holý& Šafr, 2022) ; or by incorporating partial information and allowing a compromise solution to be found between inconsistent constraints (Lenzen et al, 2006(Lenzen et al, , 2009Paelinck & Waelbroeck, 1963).…”