arXiv subjects

Yuebao Wang

Publications and source records attributed to Yuebao Wang.

17 recordsLinked to original sources

arXiv2026

Asymptotics of randomly weighted sums without moment conditions of random weights

In the paper, we investigate the asymptotics of randomly weighted sums with upper tail asymptotically independent and quasi-upper tail asymptotically independent primary random variables without requiring moment assumptions on random weights. For the case of primary random variables with regularly varying tails, we obtain more explicit results via an extension of Breiman's theorem. Then an application of the obtained results is established to asymptotically estimate for the finite-time and infinite-time ruin probabilities in a discrete-time risk model.

math.PR↗

arXiv2025

On the random-time and finite-time ruin probability for widely dependent claim sizes and inter-arrival times

Using the results of precise large deviation and renewal theory for widely dependent random variables, this paper obtains the asymptotic estimation of the random-time ruin probability and the uniform asymptotic estimation of finite-time ruin probability for a nonstandard renewal risk model, in which both claim sizes and the inter-arrival times of claim sizes are widely dependent.

math.PR↗

arXiv2024

Precise large deviations of some risk objectives related to the net loss process in two nonstandard risk models

For two nonstandard renewal risk models, we investigate the precise large deviations of the finite-time ruin probability and a random sum of the net-loss process, and the asymptotics of the random-time ruin probability. Notably, in one of these models, claim sizes series and claim interval time series are allowed to be arbitrarily dependent. Subsequently, we apply these results to obtain precise large deviations of proportional-net-loss process and excess-of-net-loss process, as well as asymptotic estimates of the mean of stop-net-loss reinsurance treaty. These results all involve the income items of the risk model, which are relatively rare in the existing references.

math.PR↗

arXiv2024

A Breiman's theorem for conditional dependent random vector and its applications to risk theory

In this paper, we give a Breiman's theorem for conditional dependent random vector, where one component has a regularly-varying-tailed distribution with the index $α\ge0$ and its slowly varying function satisfies a relaxed condition, while the other component is non-negative and its tail distribution is lighter than the former. This result substantially extends and improves Theorem 2.1 of Yang and Wang (Extremes,\ 2013). %with a lower moment condition requirement for many occasions. We also provide some concrete examples and some interesting properties of conditional dependent random vector. Further, we apply the above Breiman's theorem to risk theory, and obtain two asymptotic estimates of the finite-time ruin probability and the infinite-time ruin probability of a discrete-time risk model, in which the corresponding net loss and random discount are conditionally dependent.

math.PR↗

arXiv2021

Precise large deviations of sums of widely dependent random variables and its applications

In this paper, we obtain some results on precise large deviations for non-random and random sums of widely dependent random variables with common dominatedly varying tail distribution or consistently varying tail distribution on $(-\infty,\infty)$. Then we apply the results to reinsurance and insurance and give some asymptotic estimates on proportional reinsurance, random-time ruin probability and the finite-time ruin probability.

math.PR↗

arXiv2020

Dividend and Capital Injection Optimization with Transaction Cost for Spectrally Negative Lévy Risk Processes

For an insurance company with reserve modeled by the spectrally negative Lévy process, we study the optimal impulse dividend maximizing the expected accumulated net dividend payment subtracted by the accumulated cost of injecting capital. In this setting, the beneficiary of the dividends injects capital to ensure a non-negative risk process so that the insurer never goes bankrupt. The optimal impulse dividend and capital injection strategy together with its value function are obtained.

math.OC↗

arXiv2019

On the long tail property of product convolution

Let $X$ and $Y$ be two independent random variables with corresponding distributions $F$ and $G$ supported on $[0,\infty)$. The distribution of the product $XY$, which is called the product convolution of $F$ and $G$, is denoted by $H$. In this paper, some suitable conditions about $F $ and $G $ are given, under which the distribution $H$ belongs to the long-tailed distribution class. Here, $F$ is a generalized long-tailed distribution and is not necessarily an exponential distribution. Finally, a series of examples are given to show that the above conditions are satisfied by many distributions and one of them is necessary in some sense.

math.PR↗

arXiv2018

On the almost decrease of a subexponential density

For a subexponential density, so far, there has been no positive conclusion or counter example to show whether it is almost decreasing. In this paper, a subexponential density supported on $\mathbb{R}^+\cup\{0\}$ without the almost decrease is constructed by a little skillful method. The density is a positive piecewise linear function with a more normal shape. Correspondingly, there exists a local subexponential distribution which is not locally almost decreasing. Based on an example of Cline \cite{C1986}, some similar results are also obtained for the long-tailed density excluding the subexponential density and the local long-tailed distribution excluding the local subexponential distribution. Finally, the paper shows that, for the local subexponentiality of a distribution supported on $\mathbb{R}$, the local almost decreasing condition is necessary in some sense.

math.PR↗

arXiv2018

Asymptotics of convolution with the semi-regular-variation tail and its application to risk

In this paper, according to a certain criterion, we divide the exponential distribution class into three subclasses. One of them is closely related to the regular-variation-tailed distribution class, so it is called the semi-regular-variation-tailed distribution class. In the class, although all distributions are not convolution equivalent,they still have some good properties. We give the precise tail asymptotic expression of convolution of these distributions, and prove that the new class is closed under convolution. In addition, we do not need to require the corresponding random variables to be identically distributed. Finally, we apply these results to a discrete time risk model with stochastic returns, and obtain the precise asymptotic estimation of the finite time ruin probability.

math.PR↗

arXiv2017

A necessary and sufficient condition for the subexponentiality of product distribution

Let X and Y be two independent and nonnegative random variables with corresponding distributions F and G. Denote by H the distribution of the product XY , called the product convolution of F and G. Cline and Samorodnitsky (1994) proposed sufficient conditions for H to be subexponential, given the subexponentiality of F. Relying on a related result of Tang (2008) on the long-tail of product convolution, we obtain a necessary and sufficient condition for the subexponentiality of H, given that of F. We also study the reverse problem and obtain sufficient conditions for the subexponentiality of F given that of H. Finally, we apply the obtained results to the asymptotic study of the ruin probability in a discrete-time insurance risk model with stochastic returns.

math.PR↗

arXiv2017

Some positive conclusions related to the Embrechts-Goldie' conjecture

In this paper, we give some conditions, under which, if an infinitely divisible distribution supported on $[0,\infty)$ belongs to the intersection of exponential distribution class $\mathcal{L}(γ)$ for some $γ\ge0$ and generalised subexponential distribution class $\mathcal{OS}$, then its L$\rm\acute{e}$vy spectral distribution or convolution of the distribution with itself also belongs to the same one. To this end, we discuss the closure under the compound convolution roots for the class. In addition, we do some in-depth discussion about the above-mentioned conditions, and provide some types of distributions satisfying them. Further, we obtain some local versions of the above-mentioned results by the Esscher transform of distributions. Therefore, some positive conclusions related to the Embrechts-Goldie conjecture are obtained. Prior to this, all corresponding results are negative

math.PR↗

arXiv2017

The uniform local asymptotics of the total net loss process in a new time-dependent bidimensional renewal model

In this paper, we consider a bidimensional renewal risk model with constant force of interest, in which the claim size vector with certain local subexponential marginal distribution and its inter-arrival time are subject to a new time-dependence structure. We obtain the uniform local asymptotics of the total net loss process in the model. Moreover, some specific examples of the joint distribution satisfying the conditions of the dependence structure are given. Finally, in order to illustrate a condition of the above result, a local subexponential distribution is find for the first time that, its local distribution is not almost decreased.

math.PR↗

arXiv2016

On closedness under convolution roots related to an infinitely divisible distribution in the distribution class L(γ)

We consider questions related to the well-known conjecture due to Embrechts and Goldie on the closedness of different classes of heavy- and light-tailed distributions with respect to convolution roots. We show that the class L(γ)\cap OS is not closed under convolution roots related to an infinitely divisible distribution for any γ\ge0, i.e. we provide examples of infinitely divisible distributions belonging to this class such that the corresponding Levy spectral distribution does not. We also prove a similar statement for the class (L(γ)\cap OS)\ S(γ). In order to facilitate our analysis, we explore the structural properties of some of the classes of distributions, and study some properties of the well-known transformation from a heavy-tailed distribution to a light-tailed one.

math.PR↗

arXiv2016

The local asymptotic estimation for the supremum of a random walk with generalized strong subexponential summands

In this paper, the local asymptotic estimation for the supremum of a random walk and its applications are presented. The summands of the random walk have common long-tailed and generalized strong subexponential distribution. This distribution class and the corresponding generalized local subexponential distribution class are two new distribution classes with some good properties. Further, some long-tailed distributions with intuitive and concrete forms are found, which show that the intersection of the two above-mentioned distribution classes with long-tailed distribution class properly contain the strong subexponential distribution class and the locally subexponential distribution class, respectively.

math.PR↗

arXiv2015

Convolution and convolution-root properties of long-tailed distributions

We obtain a number of new general properties, related to the closedness of the class of long-tailed distributions under convolutions, that are of interest themselves and may be applied in many models that deal with "plus" and/or "max" operations on heavy-tailed random variables. We analyse the closedness property under convolution roots for these distributions. Namely, we introduce two classes of heavy-tailed distributions that are not long-tailed and study their properties. These examples help to provide further insights and, in particular, to show that the properties to be both long-tailed and so-called "generalised subexponential" are not preserved under the convolution roots. This leads to a negative answer to a conjecture of Embrechts and Goldie [10, 12] for the class of long-tailed and generalised subexponential distributions. In particular, our examples show that the following is possible: an infinitely divisible distribution belongs to both classes, while its Levy measure is neither long-tailed nor generalised subexponential.

math.PR↗

arXiv2015

On the structure of a class of distributions obeying the principle of a single big jump

In this paper, we present several heavy-tailed distributions belonging to the new class J of distributions obeying the principle of a single big jump introduced by Beck et al. [1]. We describe the structure of this class from different angles. First, we show that heavy-tailed distributions in the class J are automatically strongly heavy-tailed and thus have tails which are not too irregular. Second, we show that such distributions are not necessarily weakly tail equivalent to a subexponential distribution. We also show that the class of heavy-tailed distributions in J which are neither long-tailed nor dominatedly-varying-tailed is not only non-empty but even quite rich in the sense that it has a nonempty intersection with several other well-established classes. In addition, the integrated tail distribution of some particular of these distributions shows that the Pakes-Veraverbeke-Embrechts Theorem for the class J in [1] does not hold trivially.

math.PR↗

arXiv2014

On a transformation between distributions obeying the principle of a single big jump

Beck et al. (2013) introduced a new distribution class J which contains many heavy-tailed and light-tailed distributions obeying the principle of a single big jump. Using a simple transformation which maps heavy-tailed distributions to light-tailed ones, we find some light-tailed distributions, which belong to the class J but do not belong to the convolution equivalent distribution class and which are not even weakly tail equivalent to any convolution equivalent distribution. This fact helps to understand the structure of the light-tailed distributions in the class J and leads to a negative answer to an open question raised by the above paper.

math.PR↗