An algorithm for moment-matching scenario generation with application to financial portfolio optimisation

Abstract

We present an algorithm for moment-matching scenario generation. This method produces scenarios and corresponding probability weights that match exactly the given mean, the covariance matrix, the average of the marginal skewness and the average of the marginal kurtosis of each individual component of a random vector. Optimisation is not employed in the scenario generation process and thus the method is computationally more advantageous than previous approaches. The algorithm is used for generating scenarios in a mean-CVaR portfolio optimisation model. For the chosen optimization example, it is shown that desirable properties for a scenario generator are satisfied, including in-sample and out-of-sample stability. It is also shown that optimal solutions vary only marginally with increasing number of scenarios in this example; thus, good solutions can apparently be obtained with a relatively small number of scenarios. The proposed method can be used either on its own as a computationally inexpensive scenario generator or as a starting point for non-convex optimisation based scenario generators which aim to match all the third and the fourth order marginal moments (rather than average marginal moments).

Development of an inexact risk-aversion optimization model for regional carbon constrained electricity system planning under uncertainty

“…Based on a series of analysis, it was found that the distribution of CVaR is uniformed, thus, in this paper, a is the considered equal to 0.95 [34,35]. In the model, k is a nonnegative trade-off coefficient representing the exchange rate of mean cost for risk, which is specified by decision makers according to their risk preferences, and it can be chosen as any real numbers.…”

“…where k is the weighting factor presenting the tradeoff between the expected cost and CVaR; n ± denotes the value of VaR; a is confidence level, usually it could be set as 0.90, 0.95 and 0.99 [34,35]. subject to:…”

Multivariate dependence risk and portfolio optimization: An application to mining stock portfolios

“…Besides, it exposes the difficulties involved in the optimization of portfolios and the limitations of mathematical models to account for all the information and sources of risk. Each of the risk measures applied in this study differs from the others by stressing a particular portfolio optimization component, as opposed to various methodologies already presented in the recent relevant literature, such as in Hassanzadeh et al (2014), Levy and Levy (2014), Utz et al (2014) or Ponomareva et al, (2015). For instance, in our work the CVaR and CDaR measures stress the modeling of returns values falling below a threshold value commonly expressed as a horizontal line (i.e.…”

The Scenario-Based Approach to Trade in Option Contracts

“…To eliminate the controversy, the author suggest setting two or three macroscenarios, which would subsequently engender their individuals scenarios, thus ensuring the logic consistency of the tree. The studies exemplify how scenario trees are made [16][17][18].…”