* If you want to update the article please login/register

Under the two-asset Kou jump-diffusion scheme, this paper addresses the numerical implementation of the two-dimensional time-dependent partial integro-differential equation that holds for the values of European-style options. The two-dimensional Kou integral's computational complexity has the lowest computational cost: the number of basic arithmetic operations is directly proportional to the number of spatial grid points in the semidiscretization. We investigate seven contemporary operator splitting schemes of the implicit and the alternating direction implicit kinds for the successful discretization in time.

Source link: https://arxiv.org/abs/2207.10060v1

This service is characterized as the geometric mean of the agents' assessments and, as a result, it strikes a balance between the extremes of social security and egalitarian welfare, resulting in a compromise between the extremes of social justice and egalitarian concern. In recent years, Nash social wellbeing has been extensively researched for various valuation classes. In particular, a notable negative result is known when the agents' valuations are comparable and reported by value queries: for XOS valuations, one must perform exponentially many value queries to determine any substantial approximation for NSW. We use demand oracles and XOS oracles in this regard; both of these query techniques are common and have been used in previous work on social justice maximization with XOS valuations. We're developing the first subestimation algorithm for maximizing Nash social security under XOS valuations, which can be specified by demand and XOS oracles.

Source link: https://arxiv.org/abs/2110.00767v2

We investigate fair allocation of indivisible products when agents have matroid rank values. Our solution is scalable; it is provably faster than all known algorithms to calculate Lorenz dominating allocations; it is provably quicker than most existing algorithms to compute Lorenz dominating allocations. In any true fair allocation setting, algorithms can be adopted by algorithms; our contribution brings us one step closer to this goal. These two properties are crucial to the adoption of algorithms in any real fair allocation scheme; any real fair allocation setting will have these two properties.

Source link: https://arxiv.org/abs/2206.08495v4

Option pricing models, hedging algorithms, and pricing models of equity-linked life insurance products have been developed for a public company. This paper presents a log private company valuation model, which is based on the popular Gordon growth model for this reason. For a private company, we find closed-form option pricing formulas, hedging formulas, and net premium formulas for equity-linked life insurance products in this paper.

Source link: https://arxiv.org/abs/2206.09666v2

They found that a steady hyperbolic curve covering a discrete valuation field has steady decline, if and only if the Jacobian variety of the curve is stable reduction in the case where the curve's residue field is algebraically closed. We present a model of minimal log regular curve models in this paper.

Source link: https://arxiv.org/abs/2205.11964v4

We find that the inherent randomness of the widely used stochastic gradient descent might lead to inconsistent data value estimates across different runs. We show that the Banzhaf value, a value concept that stems from cooperative game theory research, has the highest resilience among all semivalues, a group of value notions that meet requisite properties entailed by ML applications. We derive the lower bound sample complexity for Banzhaf value approximation, and we report that our MSR algorithm's sample size almost matches the lower bound. Overall, our report shows that when the underlying ML algorithm is stochastic, the Banzhaf value is a promising alternative to semivalue-based data value schemes given its computational convenience and ability to consistently distinguish data quality.

Source link: https://arxiv.org/abs/2205.15466v3

We show that the value group and the ordered residue field of an ordered hensely valued field are stably embedded, proving the latter.

Source link: https://arxiv.org/abs/2206.15301v1

Fair division of indivisible items is a well-studied topic in Economics and Computer Science, where each agent has a value for each subset of items. We show that a complete EFX allocation exists even when each agent is of one of two given types, where agents of the same origin have similar additive values. This is the first time an agent and items can be specified for individual items in non-identical valuations, where there are a number of agents and products, but there is no limit on how many different values an agent can have for individual items. We provide a concrete example, in which we successfully obtain a Pareto dominating EFX allocation from an existing partial EFX allocation.

Source link: https://arxiv.org/abs/2008.08798v3

For any polarized variety, we show that test setups and, more generally, R-test configurations can be investigated in terms of Fubini-Study functions on the Berkovich analytization of X with respect to its trivial absolute value on the ground field.

Source link: https://arxiv.org/abs/2107.11221v2

* Please keep in mind that all text is summarized by machine, we do not bear any responsibility, and you should always check original source before taking any actions