A Non-Sign-Preserving Ras Variant

Abstract

We have developed a variant of the RAS generalised iterative scaling method that is able to change the sign between successive iterates, and thus fulfil constraints that are infeasible for existing RAS variants. Like earlier RAS variants, our method can handle constraints on arbitrarily sized and shaped subsets of matrix elements, include reliability of the initial estimate and the external constraints, and deal with negative values.

Calculation of growth contributions using structural decomposition of input-output tables

“…7 Az input-output táblázatok továbbvezetésére és kiegyensúlyozására a klasszikus RAS-eljáráson kívül lineáris és kvadratikus programozási technikák is használatosak (lásd például Lahr-Mesnard [2004] vagy Jackson-Murray [2004]), illetve a RAS-nak is léteznek különféle (nem előjelmegőrző, nulla és negatív peremekre is jól működő) változatai (lásd például Lenzen et al [2014] tanulmányát vagy Révész [2001] "additív" RAStechnikáját). Egyszerűsége miatt azonban még mindig a hagyományos RAS a legelterjedtebb kiegyenlítési eljárás.…”

A General Nonlinear Least Squares Data Reconciliation and Estimation Method for Material Flow Analysis

“…Further developments of the RAS method include MRAS, where some transfer coefficients are known with certainty, and KRAS, the most developed of the RAS family of algorithms. The RAS family of algorithms, particularly KRAS, is well suited to MFA data reconciliation problems where data are available as a set of process values, an estimated set of either TCs or flow data, any external linear constraints on process values, as documented by Lenzen and colleagues (), and some data quality measure. However, the inclusion of nonlinear data constraints is precluded, and a user must choose between using TC or flow data for the reconciliation.…”

Updating Input–Output Tables with Benchmark Table Series

“…Jackson and Murray (2004) Kuroda's method Kuroda (1988) and Wilcoxen (1989) Cross entropy method (CE) Golan et al (1994) TRAS St Louis (1999, 2004) Information in addition to row and column totals KRAS Lenzen et al (2006Lenzen et al ( , 2007Lenzen et al ( , 2009), Wood (2011) A non-sign-preserving RAS variant Lenzen et al (2014) Cycling method for constructing I-O table time series Lenzen et al (2012) Incomplete data this paper are as follows. First, time series analysis is used to forecast I-O tables for the target years.…”

“…Wood (2011) documented the development of a series of Australian I-O tables using KRAS. Lenzen et al (2014) also developed a variant of the RAS generalized iterative scaling method that is able to change signs between successive iterations and thus satisfies signchanging constraint that existing RAS variants could not.…”