Updating input–output matrices: assessing alternatives through simulation

A problem that frequently arises in economics, demography, statistics, transportation planning and stochastic modelling is how to adjust the entries of a matrix to fulfil row and column aggregation constraints. Biproportional methods in general and the so-called RAS algorithm in particular, have been used for decades to find solutions to this type of problem. Although alternatives exist, the RAS algorithm and its extensions are still the most popular. Apart from some interesting empirical and theoretical properties, tradition, simplicity and very low computational costs are among the reasons behind the great success of RAS. Nowadays computer hardware and software have made alternative procedures equally attractive. This work analyses, through simulation, the performance of RAS and some minimands when matrix coefficients vary following different schemes of change. Results suggest RAS algorithm as the best option when variations in coefficients are proportional to their size, while the method based on minimizing squared differences is seen to be the best alternative when the standard deviations of variations are either constant, variable, or an inverse function of matrix entries.

[1]  W. Deming,et al.  On a Method of Estimating Birth and Death Rates and the Extent of Registration (Excerpt) , 1949 .

[2]  M. Fréchet,et al.  Sur les tableaux dont les marges et des bornes sont données , 1960 .

[3]  R. Stone,et al.  Input-output and national accounts , 1961 .

[4]  Richard Stone,et al.  A Programme for Growth, 1. A Computable Model of Economic Growth , 1963 .

[5]  T. I. Matuszewski,et al.  Linear Programming Estimates of Changes in Input Coefficients , 1964 .

[6]  Michael Bacharach,et al.  Biproportional matrices & input-output change , 1970 .

[7]  Michael Green.  Biproportional Matrices and Input‐Output Change , 1971 .

[8]  J. Darroch,et al.  Generalized Iterative Scaling for Log-Linear Models , 1972 .

[9]  Joseph E. Haring,et al.  AN APPRAISAL OF NONSURVEY TECHNIQUES FOR ESTIMATING REGIONAL INPUT‐OUTPUT MODELS* , 1974 .

[10]  W. F. Gossling,et al.  Estimating and projecting input output coefficients , 1975 .

[11]  Lenclud,et al.  Ajustement de tableaux statistiques , 1976 .

[12]  Thionet.  Construction et reconstruction de tableaux statistiques , 1976 .

[13]  William H. Miernyk,et al.  Comments on Recent Developments in Regional Input-Output Analysis , 1976 .

[14]  R C Hinojosa.  A Performance Test of the Biproportional Adjustment of Input—Output Coefficients , 1978 .

[15]  Philip Rees,et al.  Regional population projection models and accounting methods , 1979 .

[16]  R. Krupp.  Properties of krulthof's Projection method , 1979, The Bell System Technical Journal.

[17]  N. F. Stewart,et al.  Bregman's balancing method , 1981 .

[18]  Malachy Carey,et al.  A Method for Direct Estimation of Origin/Destination Trip Matrices , 1981 .

[19]  Ron Johnston,et al.  Estimating the Sources of Spatial Change in Election Results: A Multiproportional Matrix Approach , 1982 .

[20]  G. Kusic,et al.  Tie Line Contingency Studies Based Upon Partial Information , 1983, IEEE Transactions on Power Apparatus and Systems.

[21]  Karen R. Polenske,et al.  A LINEAR PROGRAMMING APPROACH TO SOLVING INFEASIBLE RAS PROBLEMS , 1987 .

[22]  John M. Mulvey,et al.  Balancing large social accounting matrices with nonlinear network programming , 1989, Networks.

[23]  Bhat Pn,et al.  Estimating transition probabilities of age misstatement. , 1990 .

[24]  Stavros A. Zenios,et al.  A Comparative Study of Algorithms for Matrix Balancing , 1990, Oper. Res..

[25]  Ron Johnston,et al.  Entropy-Maximizing and the Iterative Proportional Fitting Procedure , 1993 .

[26]  Albert E. Steenge,et al.  Input-Output and National Accounts , 1993 .

[27]  F. Willekens,et al.  Monitoring international migration flows in Europe , 1994, European journal of population = Revue europeenne de demographie.

[28]  Y. Vardi,et al.  Network Tomography: Estimating Source-Destination Traffic Intensities from Link Data , 1996 .

[29]  H. Meyer,et al.  Diagonal scaling and the analysis of polarization experiments in nuclear physics , 1997 .

[30]  Michel Balinski,et al.  Mexican Electoral Law: 1996 Version , 1997 .

[31]  Karen R. Polenske,et al.  Current Uses of the RAS Technique: A Critical Review , 1997 .

[32]  Paola Sebastiani,et al.  Parameter Estimation in Bayesian Networks from Incomplete Databases , 1998, Intell. Data Anal..

[33]  Robert McDougall.  Entropy Theory and RAS are Friends , 1999 .

[34]  Ali Reza Jalili,et al.  Evaluating Relative Performances of Four Non-Survey Techniques of Updating Input-Output Coefficients , 2000 .

[35]  M. Carapeto,et al.  On model complexity and selection , 2003 .

[36]  Robert Schoen,et al.  Estimating multistate transition rates from population distributions , 2003 .

[37]  Alan T. Murray,et al.  Alternative Input-Output Matrix Updating Formulations , 2004 .

[38]  Louis de Mesnard,et al.  Biproportional Techniques in Input-Output Analysis: Table Updating and Structural Analysis , 2004 .

[39]  Jérôme Pagès,et al.  A principal axes method for comparing contingency tables: MFACT , 2004, Comput. Stat. Data Anal..

[40]  Michael L. Lahr,et al.  A Strategy for Producing Hybrid Regional Input-Output Tables , 2005 .

[41]  István Varga,et al.  ESTIMATION OF DYNAMIC ORIGIN DESTINATION MATRIX OF TRAFFIC SYSTEMS , 2005 .

[42]  P. Canning,et al.  A Flexible Mathematical Programming Model to Estimate Interregional Input-Output Accounts , 2005 .

[43]  Satoshi Aoki,et al.  Markov chain Monte Carlo exact tests for incomplete two-way contingency tables , 2005 .

[44]  Louis de Mesnard,et al.  A Note on Added Information in the Ras Procedure: Reexamination of Some Evidence , 2006 .

[45]  Eric O'N. Fisher,et al.  The Structure of the American Economy , 2008 .

[46]  Bahman Kalantari,et al.  On the complexity of general matrix scaling and entropy minimization via the RAS algorithm , 2007, Math. Program..

[47]  P. N. Mari Bhat,et al.  Estimating transition probabilities of age misstatement , 2010, Demography.