Balancing large systems of national accounts

  • F. Van Der Ploeg


Systems of national accounts typically contain measurement errors. Official statisticians are prepared to give initial estimates with subjective information on the standard deviations of the measurement errors of these estimates. The systems should satisfy certain accountaing identities, that is, the total incomings into any account should match the total outgoings of the account. A least-squares procedure for the adjustment of large systems of national accounts is proposed, which adjusts the initial estimates so that they satisfy the accounting restrictions. The computational requirements of this adjustment procedure are enormous as the systems of national accounts are frequently very large. Hence, the problem is reformulated and a conjugate gradient algorithm which exploits the constraint sparcity and the specific structure of economic accounting matrices is proposed. The associated computer program is self-contained, relatively cheap, and has been applied to systems of up to 262 accounts.

Key words

National accounts residual errors data inconsistency least-squares adjustment conjugate gradient algorithm 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Bacharach, M. (1970) Biproportional Matrices and Input-Output Change, CUP, Cambridge.Google Scholar
  2. Barker, T. S., Ploeg, F. van der, and Weale, M. R. (1982) Towards a balanced system of national accounts: UK 1975, CGP Discussion Paper No. 518, Cambridge University: (presented to the CSO).Google Scholar
  3. Barker, T. S., Ploeg, F. van der, and Weale, M. R. (1984) A balanced system of national accounts for the United Kingdom, Review of Income and Wealth 30, 461–485.Google Scholar
  4. Byron, R. P. (1978) The estimation of large social account matrices, J. R. Statist. Soc. A 141, 359–367.Google Scholar
  5. Dongarra, J. J. and Hewitt, T. (1986) Implementing dense linear algebra algorithms using multitasking on the GRAY X-MP-4 (or approaching the Gigaflop), SIAM Journal on Scientific and Statistical Computation 7, 347–350.Google Scholar
  6. Hestenes, M. R. and Stiefel, E. (1952) Methods of conjugate gradients for solving their systems, Res. National Bureau of Standards 49, 409–436.Google Scholar
  7. Maurice, R. (1968) National Accounts: Statistics, Sources and Methods, HMSO London.Google Scholar
  8. Ploeg, F. van der (1982) Reliability and the adjustment of sequences of large economic accounting matrices (with discussion), J. R. Statist. Soc. A 145, 169–194.Google Scholar
  9. Ploeg, F. van der (1984) Generalized least squares methods for balancing large systems and tables of national accounts, Review of Public Data Use 12, 17–33.Google Scholar
  10. Stone, J. R. N., Champernowne, D. G., and Meade, J. E. (1942) The precision of national income estimates, Rev. Econ. Studies 9, 111–125.Google Scholar
  11. U.N., Statistical Office (1968) A System of National Accounts, Studies in Methods, Series F, No. 2, rev. 3. U.N., New York.Google Scholar
  12. Wolfe, M. A. (1978). Numerical Methods for Unconstrained Optimization. Van Nostrand Reinhold, London.Google Scholar

Copyright information

© Kluwer Academic Publishers 1988

Authors and Affiliations

  • F. Van Der Ploeg
    • 1
    • 2
  1. 1.London School of EconomicsLondonUK
  2. 2.Tilburg UniversityLE TilburgThe Netherlands

Personalised recommendations