China
Africa
Explore chapters and articles related to this topic
Optimization algorithms for multiple-asset portfolios with machine learning techniques
Published in Noura Metawa, M. Kabir Hassan, Saad Metawa, Artificial Intelligence and Big Data for Financial Risk Management, 2023
In order to illustrate the composition of coherent portfolios [1,2,3,4], Tables 12.6–12.9 show the asset allocations in addition to their expected returns and systematic risk factors. Similarly, the four tables depict the minimum economic capital required to sustain the operations of these portfolios and the ratio of Economic Capital /Volume with the use of three different correlation parameters. In this way, portfolio managers should employ risk measures, which allow them to take decisions that would produce a risk budget lower than a specific target. Thus, this portfolio analysis is substantially a key improvement to the conventional Markowitz (1959) technique as our modeling algorithms and optimization techniques permit the determination of the asymmetric aspect of risk under different market outlooks. In any case, the benefit of portfolio optimization critically depends on how accurately the implemented economic capital risk measure is forecasted.
Fundamentals of Hazard, Exposure, and Risk Assessment
Published in D. Kofi Asante-Duah, Hazardous Waste Risk Assessment, 2021
Risk measures give an indication of the probability and severity of adverse effects (to health, environment, or property), and generally are established with varying degrees of confidence according to the importance of the decision involved. Risk estimation involves an integration of information on the intensity, frequency, and duration of exposure for all identified exposure routes for the exposed or affected group(s). Measures used in risk assessment assume various forms, depending on the type of problem, degree of resolution appropriate for the situation on hand, and the analysts’ preference. It may be expressed in quantitative terms, in which case it takes on values from zero (associated with certainty for no- adverse effects) to unity (associated with certainty for adverse effects to occur); in several other cases, risk is only described qualitatively, by use of descriptors such as “high,” “moderate,” “low,” etc.; or indeed, risk may be described in semiquantitative/semiqualitative terms.
Theoretical and practical foundations of liquidity-adjusted value-at-risk (LVaR)
Published in Noura Metawa, Mohamed Elhoseny, Aboul Ella Hassanien, M. Kabir Hassan, Expert Systems in Finance, 2019
In this way, portfolio managers can employ appropriate downside-risk measures that allow them to take prompt decisions, which could produce a risk budget lower than a specific target. In this line of reasoning, under adverse market conditions for instance, LVaR is calculated by implementing downside volatilities (i.e. maximum negative assets returns throughout the sampling period). Thus, this analysis and the implemented LVaR algorithm permit one to determine the asymmetric aspect of risk and investable portfolios under adverse market conditions and, as such, it is a substantial improvement to the traditional Markowitz’s MV approach.
Analysis of human-factor-caused freight train accidents in the United States
Published in Journal of Transportation Safety & Security, 2021
Zhipeng Zhang, Tejashree Turla, Xiang Liu
The limitation of using mean as the risk measure is that it fails to account for the extreme characteristics of the accidents with low probabilities but high consequences. To address this “heavy-tail” effect, prior literature has used risk measures such as value at risk (VaR) or conditional value at risk (CVaR) as alternative risk measures (Soleimani, Seyyed-Esfahani, & Kannan, 2014; Spada, Paraschiv, & Burgherr, 2018; Zhang & Liu, 2019). This study considers CVaR rather than VaR as an alternative to mean risk, in accordance with the preference of many previous studies due to CVaR’s coherency (Sarkaylin et al., 2008; Rockafellar & Uryasev, 2000). This is because VaR does not reveal anything about the magnitude of losses exceeding the VaR limit which can be addressed by CVaR. It is the weighted average of all outcomes exceeding the confidence interval of a data set sorted from worst to best. Simply put, the CVaR of the collision risk is the average of severity (number of casualties or derailed cars) of all the collisions that are more than (Equation 10). Usually, a confidence interval of 95% is adopted (Figure 7 makes a lucid representation of these measures graphically. It shows that is a chosen quantile, so the frequency at is the VaR and the average of events exceeding it is CVaR.
Risk-averse flexible policy on ambulance allocation in humanitarian operations under uncertainty
Published in International Journal of Production Research, 2021
Guodong Yu, Aijun Liu, Huiping Sun
Propositions (3) and (4) caution that the waiting time can be longer with the increase of the number of calls. Thus, to further improve the operational performance, we provide following strengthened risk-averse constraints on the value function. Risk measures (e.g, value-at-risk (VaR), conditional value-at-risk (CVaR) and mean-variance) are the most popular way of representing the risk-averseness (Rockafellar and Uryasev 2002). However, under dynamic environment, VaR and CVaR fail to satisfy a property of additive consistency, which means that the risk preference between two random variables do not change by adding another independent variable to both variables (Rockafellar and Uryasev 2000, 2002). In this paper, we will adopt the first-order stochastic dominance (FSD) as constraints to manage the risk. Although FSD leads to a non-convex optimisation problem, we can develop an equivalent linear programming to solve the problem very well.
Pricing real options based on linear loss functions and conditional value at risk
Published in The Engineering Economist, 2020
The option value determined by the linear loss function approach simply sets the upper bound on how much to pay for an option premium. To determine the appropriate amount to pay for an actual option premium, we may introduce the Conditional Value at Risk concept. The Value at Risk (VaR) is one of the well-known risk measures used by financial institutions to capture the potential loss in value of their traded portfolios from market movements over a specified period. Once the potential loss amount is determined, then this can be compared to their available capital and cash reserves to see if the losses can be covered without putting the firms at risk. When the concept is applied to non-financial firms, the Net Present Value at Risk (NPVaR) represents the worst outcome at a given confidence level on the NPV distribution. In this section, we propose to adopt the expected loss amount obtained from the Conditional Net Present Value at Risk (CNPVaR) as a basis to determine what the appropriate price to pay for option premium.