Journal of Asset Management

, Volume 20, Issue 6, pp 469–475 | Cite as

A convergence-speed-dependent data quantity definition and its effect on risk estimation

  • Jakob KrauseEmail author
Original Article


Data quantity plays a crucial role in the estimation of risk measures since ‘more data’ lead to a better estimator convergence and consequently to a better risk assessment. The objective of this paper is to use this relationship between data quantity and estimator convergence to formally derive a measure of data quantity for estimators based on weighted observations. For the case of a variance estimation and using exponentially weighted observations, this procedure leads to analytical formulas for the implied measure of data quantity. As such, this paper specifies the theoretical underpinnings of measures of data quantity which have been present in the literature (effective number of scenarios) and, as an application, demonstrates the effect of the specific measure of data quantity on risk assessment.


Data Quantity EWMA Weighted estimation 



  1. Giacomini, R., and B. Rossi. 2009. Detecting and predicting forecast breakdowns. The Review of Economic Studies 76 (2): 669–705.CrossRefGoogle Scholar
  2. Inoue, A., L. Jin, and B. Rossi. 2017. Rolling window selection for out-of-sample forecasting with time-varying parameters. Journal of Econometrics 196 (1): 55–67.CrossRefGoogle Scholar
  3. Krause, J. 2019. Representativeness vs Convergence: Optimal Data Identification in Non-Stationary Systems Working Paper.Google Scholar
  4. Meucci, A. 2012. Effective Number of Scenarios in Fully Flexible Probabilities. GARP Risk Professional, pp. 32–35, February 2012. Available at SSRN: Accessed 22 Mar 2019.
  5. Pesaran, M.H., and A. Timmermann. 2007. Selection of estimation window in the presence of breaks. Journal of Econometrics 137 (1): 134–161.CrossRefGoogle Scholar
  6. Shao, J. 1998. Mathematical Statistics. New York: Springer.Google Scholar
  7. Stock, J.H., and M.W. Watson. 1996. Evidence on structural instability in macroeconomic time series relations. Journal of Business & Economic Statistics 14 (1): 11–30.Google Scholar

Copyright information

© Springer Nature Limited 2019

Authors and Affiliations

  1. 1.School of Economics and BusinessMartin-Luther University Halle-WittenbergHalleGermany
  2. 2.European Commodity ClearingLeipzigGermany

Personalised recommendations