Answers are grounded in this paper and cite their sources.
This research estimates joint cumulants in financial data, revealing their relationships with other estimators.
As conventional realized third and fourth (co)moments estimated with sub-period returns are biased, Neuberger (2012) and Bae and Lee (2021) develop new unbiased realized (joint) cumulants using extended information. In this paper, we discuss practical issues in estimating the realized (joint) cumulants. In addition, we estimate (joint) cumulants through various methods using simulated prices and examine the characteristics of those estimators. The simulation results show that realized (joint) cumulants estimated from sub-period returns and option data serve as proxies when the true realized cumulants are not obtainable. Lastly, we estimate realized (joint) cumulants using financial data on the S&P 500, individual stocks, and their options, and investigate whether the realized (joint) cumulants are explained by other estimators. As a result, we find that realized coskewness and kurtosis are predicted by implied moments and other lagged ex post estimators.
No takes yet. Share an insight, caveat, or question.
Bae et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: