Problems in evaluating accuracy and consistency of macroeconomic forecasts

  • Received March 2, 2018;
    Accepted April 18, 2018;
    Published May 4, 2018
  • Author(s)
  • DOI
  • Article Info
    Volume 16 2018, Issue #2, pp. 102-112
  • Cited by
    2 articles

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License

The paper focuses on the analysis of the quality of macroeconomic forecasts, as well as on general issues of the functioning of the system of macroeconomic forecasting in the Republic of Kazakhstan. It provides a comparative analysis of errors of initial and revised macroeconomic forecasts of the real sector of economy indicators. To assess forecast accuracy, the authors use the world’s most widespread measures, such as the mean error and the mean absolute percentage error, as well as official statistical data and records. The overall forecast consistency and congruence of assumptions on the dynamics of indicators of the real sector and fiscal policy are estimated using the basic identities of the System of National Accounts. The paper also considers institutional aspects of the system of macroeconomic forecasting in the Republic of Kazakhstan. The authors conclude that revised forecasts, with the exception of monetary policy indicators, exhibit a smaller error, which can indicate a need for greater coordination of public authorities in the process of preparing forecasts, developing the system of independent assessment of their quality and improving their transparency.

view full abstract hide full abstract
    • Figure 1. Forecast and actual values of real GDP growth, %
    • Table 1. Errors in the forecast of the real sector of economy, pp.
    • Table 2. Errors in the forecast of monetary statistics indicators, pp.
    • Table 3. Errors in the forecast of the government budget indicators, pp.
    • Table 4. Correlation coefficients of forecast errors for 2002–2016
    • Table 5. Forecast indicators for 2018
    • Table 6. Indicators of net lending/borrowing, trillion tenge