Conditional Value-at-Risk for Uncommon Distributions

Khokhlov, Valentyn

doi:10.2139/ssrn.3200629

Cited by 6 publications

(7 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The control and experimental groups are not applicable in our discussion because of openly observed (not experimental) collection of COVID-19. The risk 3 of contracting the COVID-19 virus during a cruise is more than in a community setting as confined spaces discourage non-pharmaceutical mitigation strategies such as social distancing, to be weakly implemented and, not to mention, breathing air is tightly internalized. More nations are afraid to let the voyagers come ashore at the seaports.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Stochasticity among Victims of COVID-19 Pandemic

Shanmugam¹,

Ledlow²,

Singh³

2022

JMDH

View full text Add to dashboard Cite

This article provides a thorough explanation of methods and theoretical concepts to detect infectivity of COVID-19. The concept of heterogeneity is discussed and its impacts on COVID-19 pandemics are explored. Observable heterogeneity is distinguished from nonobservable heterogeneity. The data support the concepts of heterogeneity and the methods to extract and interpret the data evidence for the conclusions in this paper. Heterogeneity among the vulnerable to COVID-19 is a significant factor in the contagion of COVID-19, as demonstrated with incidence rates using data of a Diamond Princess cruise ship. Given the nature of the pandemic, its heterogeneity with different social norms, pre-and post-voyage quick testing procedures ought to become the new standard for cruise ship passengers and crew. With quick testing, identification of those infected and thus, not allowing to embark on a cruise or quarantine those disembarking, and other mitigation strategies, the popular cruise adventure could become norm for safe voyage. The novel method used in this article adds valuable insight in the modeling of disease and specifically, the COVID-19 virus.

show abstract

Section: Discussionmentioning

confidence: 99%

“…More nations are afraid to let the voyagers come ashore at the seaports. 3 Not even the ships are permitted to dock at the port as to not complicate virus mitigation efforts by the local surrounding communities. The scenario seems to be anti-humanistic.…”

Section: Discussionmentioning

confidence: 99%

Stochasticity among Victims of COVID-19 Pandemic

Shanmugam¹,

Ledlow²,

Singh³

2022

JMDH

View full text Add to dashboard Cite

show abstract

“…Given the Gaussian approximation, the CVaR measure is easily computed (Khokhlov, 2016;Tang et al, 2020). At each iteration, Q c (s, a) and V c (s, a) can be estimated.…”

Section: Gaussian Approximationmentioning

confidence: 99%

“…At each iteration, Q c (s, a) and V c (s, a) can be estimated. Thus, the new safety measure for risk level is computed by where (⋅) and Φ(⋅) denote the probability distribution function (PDF) and the cumulative distribution function (CDF) of the standard normal distribution (Khokhlov, 2016).…”

Section: Gaussian Approximationmentioning

confidence: 99%

Safety-constrained reinforcement learning with a distributional safety critic

et al. 2022

View full text Add to dashboard Cite

Safety is critical to broadening the real-world use of reinforcement learning. Modeling the safety aspects using a safety-cost signal separate from the reward and bounding the expected safety-cost is becoming standard practice, since it avoids the problem of finding a good balance between safety and performance. However, it can be risky to set constraints only on the expectation neglecting the tail of the distribution, which might have prohibitively large values. In this paper, we propose a method called Worst-Case Soft Actor Critic for safe RL that approximates the distribution of accumulated safety-costs to achieve risk control. More specifically, a certain level of conditional Value-at-Risk from the distribution is regarded as a safety constraint, which guides the change of adaptive safety weights to achieve a trade-off between reward and safety. As a result, we can compute policies whose worst-case performance satisfies the constraints. We investigate two ways to estimate the safety-cost distribution, namely a Gaussian approximation and a quantile regression algorithm. On the one hand, the Gaussian approximation is simple and easy to implement, but may underestimate the safety cost, on the other hand, the quantile regression leads to a more conservative behavior. The empirical analysis shows that the quantile regression method achieves excellent results in complex safety-constrained environments, showing good risk control.

show abstract

“…The hazard rate to contract COVID-19 in a situation in which the family union consists of attendees having heterogeneous resistance to COVID-19 is The reduction in the hazards due to having attendees with heterogeneously resistant to COVID-19 in the family union is . We next consider a popular concept called Tail Value at Risk (TVaR) in the business world(Khokhlov, 2016 for details) and it is useful in the context of contracting COVID-19. That is,…”

mentioning

confidence: 99%

Predicting COVID-19 cases with unknown homogeneous or heterogeneous resistance to infectivity

Shanmugam

Ledlow

Singh

2020

Preprint

View full text Add to dashboard Cite

This article constructs a restricted infection rate inverse binomial-based approach to predict COVID-19 cases after a family gathering. The traditional inverse binomial (IB) model is unqualified to match the reality of COVID-19, because the data contradicts the model’s requirement that variance should be greater than expected value. A refined version of the IB model is a necessity to predict COVID-19 cases after family gatherings. Our refined version of an IB model is more appropriate and versatile, as it accommodates all potential data scenarios: equal, lesser, or greater variance than expected value.Application of the approach is based on a restricted infectivity rate and methodology on Fan et al.’s COVID-19 data, which exhibits two clusters of infectivity. Cluster 1 has a smaller number of primary cases and exhibits larger variance than the expected cases with a negative correlation of 28%, implying that the number of secondary cases is lesser when the number of primary cases increases and vice versa. The traditional inverse binomial (IB) model is appropriate for Cluster 1. The probability of contracting COVID-19 is estimated to be 0.13 among the primary, but is 0.75 among the secondary in Cluster 1, with a wider gap. Conversely, Cluster 2, exhibits smaller variance than the expected cases with a correlation of 79%, implying the number of primary and secondary cases increase or decrease together. Cluster 2 disqualifies the traditional IB model and demands its refined version. Probability of contracting COVID-19 is estimated to be 0.74 among the primary, but is 0.72 among the secondary in Cluster 2, with a narrower gap.The model’s ability to estimate the community’s health system memory for future policies to be developed is an asset of this approach. The current hazard level to be infected with COVID-19 among the primary and secondary groups are estimable and interpretable.Author SummaryCurrent statistical models are not able to accurately predict disease infection spread in the COVID-19 pandemic. We have applied a widely-used inverse binomial method to predict rates of infection after small gatherings, going from primary (original) cases to secondary (later) cases after family gatherings or social events, using the data from the Wuhan and Gansu provinces in China, where the virus first spread. The advantages of the proposed approach include that the model’s ability to estimate the community’s health system memory for future policies to be developed, as such policies might reduce COVID’s spread if not its control. In our approach, as demonstrated, the current hazard level of becoming infected with COVID-19 and the odds of contracting COVID-19 among the primary in comparison to the secondary groups are estimable and interpretable. We hope the proposed approach will be used in future epidemics.

show abstract

Conditional Value-at-Risk for Uncommon Distributions

Cited by 6 publications

References 9 publications

Stochasticity among Victims of COVID-19 Pandemic

Stochasticity among Victims of COVID-19 Pandemic

Safety-constrained reinforcement learning with a distributional safety critic

Predicting COVID-19 cases with unknown homogeneous or heterogeneous resistance to infectivity

Contact Info

Product

Resources

About