Comparison of Risk Models for Setting Margin for Futures and Option Portfolios

Calculation of margin requirements is one of the key tasks of any exchange where derivatives are traded. The amount of margin represents the collateral that ensures the reliability of transactions on the exchange by covering the exchange's losses from the need to fulfill obligations arising in the event of default of one of the agents. If the margin assigned is too small, the exchange suffers losses when agents default, as it is required to cover all losses associated with the default. If the margin is too high, the exchange loses earnings on transaction commissions, clearing, and other aspects of its activities in this area because the market becomes less liquid. The results of the margin assessment depend on the risk-parameter estimation system by which potential losses on open positions are calculated. The quality of the methodology is also a hot topic because exchange-traded derivatives - options and futures, unlike stocks and bonds, have a finite and predetermined life period. Therefore, the literature on the analysis of margin requirement valuation systems often compares several systems or theorizes the reliability of one of them as a risk-based measure. However, there has been no research on whether the assigned margin level is sufficient and exactly how the risk-measurement methodology can be adjusted in a data-poor situation to reduce marginal requirements or increase reliability. The purpose of this paper is to evaluate the reliability of the methodology for calculating the variation margin of the SPAN margin system in the Russian market and to build an alternative model that uses a larger set of random scenarios. In case the existing methodology meets the given reliability requirements, the analysis will be carried out in the area of finding a method of risk-parameter estimation, which will reduce the level of requirements without reducing the quality of the model.