BZPEERPapers, people, evidence and research pathsDiscover
Research tools
Network
Opportunities
Account
Source author record
Shiqi Song appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.
Catalog footprint
Research graph
Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.
BZPEER is loading the nearby papers, people, topics and institutions for this page.
Published work
We characterize the random times $ρ$ whose Azema supermartingales $Z^ρ$ take the form $Z^ρ=U/U^*$ for some non negative local martingales $U$ starting from 1 vanishing at infinity, where $U^*$ denotes the running maximum process of $U$.
Let $\mathbb{F}\subset \mathbb{G}$ be two filtrations and $S$ be a $\mathbb{F}$ semimartingale possessing a $\mathbb{F}$ local martingale deflator. Consider $τ$ a $\mathbb{G}$ stopping time. We study the problem whether $S^{τ-}$ or $S^τ$ can have $\mathbb{G}$ local martingale deflators. A suitable theoretical framework is set up in this paper, within which necessary/sufficient conditions for the problem to be solved have been proved. Under these conditions, we will construct $\mathbb{G}$ local martingale deflators for $S^{τ-}$ or for $S^τ$. Among others, it is proved that $\mathbb{G}$ local martingale deflators are multiples of $\mathbb{F}$ local martingale deflators, with a multiplicator coming from the multiplicative decomposition of the Azéma supermartingale of $τ$. The proofs of the necessary/sufficient conditions require various results to be established about Azéma supermartingale, about local martingale deflator, about filtration enlargement, which are interesting in themselves. Our study is based on a filtration enlargement setting. For applications, it is important to have a method to infer the existence of such setting from the knowledge of the market information. This question is discussed at the end of the paper.
When the \textit{martingale representation property} holds, we call any local martingale which realizes the representation a \textit{representation process}. There are two properties of the \textit{representation process} which can greatly facilitate the computations under the \textit{martingale representation property}. Actually, on the one hand, the \textit{representation process} is not unique and there always exists a \textit{representation process} which is locally bounded and has pathwisely orthogonal components outside of a predictable thin set. On the other hand, the jump measure of a \textit{representation process} satisfies the \textit{finite predictable constraint}. In this paper, we give a detailed account of these two properties. As application, we will prove that, under the \textit{martingale representation property}, the \textit{full viability} of an expansion of market information flow implies the \textit{drift multiplier assumption}.
It is a working version of a lecture on the theory of enlargement of filtration, given at the African Mathematic School in Marrakech, October 19-23, 2015.
We study multi-default model which satisfies the quasi-left-continuity, the martingale representation property, the drift multiplier assumption and the full viability. We use $\natural$-model to construct one such model.
A triplet $(\mathbb{P},\mathbb{F},S)$ of a probability measure $\mathbb{P}$, of an information flow $\mathbb{F}=(\mathcal{F}_t)_{t\in\mathbb{R}_+}$, and of an $\mathbb{F}$ adapted asset process $S$, is a financial market model, only if it is viable. In this paper we are concerned with the preservation of the market viability, when the information flow $\mathbb{F}$ is replaced by a bigger one $\mathbb{G}=(\mathcal{G}_t)_{t\geq 0}$ with $\mathcal{G}_t\supset\mathcal{F}_t$. Under the assumption of martingale representation property in $(\mathbb{P},\mathbb{F})$, we prove a necessary and sufficient condition for all viable market in $\mathbb{F}$ to remain viable in $\mathbb{G}$.
A supermartingale deflator (resp., local martingale deflator) multiplicatively transforms nonnegative wealth processes into supermartingales (resp., local martingales). The supermartingale numeraire (resp., local martingale numeraire) is the wealth processes whose reciprocal is a supermartingale deflator (resp., local martingale deflator). It has been established in previous literature that absence of arbitrage of the first kind (NA1) is equivalent to existence of the supermartingale numeraire, and further equivalent to existence of a strictly positive local martingale deflator; however, under NA1, the local martingale numeraire may fail to exist. In this work, we establish that, under NA1, any total-variation neighbourhood of the original probability has an equivalent probability under which the local martingale numeraire exists. This result, available previously only for single risky-asset models, is in striking resemblance with the fact that any total-variation neighbourhood of a separating measure contains an equivalent $σ$-martingale measure. The presentation of our main result is relatively self-contained, including a proof of existence of the supermartingale numeraire under NA1. We further show that, if the Levy measures of the asset-price process have finite support, NA1 is equivalent to existence of the local martingale numeraire with respect to the original probability.
Given a reference filtration $\mathbb{F}$, we develop in this work a generic method for computing the semimartingale decomposition of $\mathbb{F}$-martingales in some specific enlargements of $\mathbb{F}$. This method is then applied to the study of progressive enlargement with multiple non-ordered random times, for which explicit decompositions can be obtained under the absolute continuity condition of Jacod.
The density hypothesis on random times becomes now a standard in modeling of risks. One of the basic reasons to introduce the density hypothesis is the desire to have a computable credit risk model. However, recent work shows that merely an existence of a density function for the conditional law of the random times will not be enough for the purposes of some numerical implantation problems. It becomes necessary to have models with martingales of density functions evolving along with the development of the information flow, in particular, to have Markovian martingales of density functions determined by a stochastic differential equation. The quetion of constructing a martingale of density functions by a stochastic differential equation has been answered in one dimensional case. The aim of this note is to provide a solution in higher dimensional cases.
We consider the so-called $\natural$-model. It is an one-default model which gives the conditional law of a random time with respect to a reference filtration. This model has been studied in the case where the parameters are continuous. In this paper we will establish the $\natural$-model in the case of jump parameters. We then prove the corresponding enlargement of filtration formula and we compute the derivative of the conditional distribution functions of the random time.
In Karatzas and Kardaras's paper on semimartingale financial models, it is proved that the NUPBR condition is a property of the local characteristic of the asset process alone. In Takaoka's paper on NUPBR, it is proved that the NUPBR condition is equivalent to the existence of a simga-martingale deflator. However, Takaoka's paper founds its proof on Delbaen and Schachermayer's fundamental asset pricing theorem, i.e. the NFLVR condition, which is not a pure property of the local characteristic of the asset process. In this paper we give an alternative proof of the result of Takaoka, which makes use only the properties of the local characteristic of the asset process.
The enlargement of filtration theory is a study of semimartingales when the basic filtration changes. This theory provides particular techniques on stochastic calculus. We present here a technique, that we call the local solution method. We will show, with several examples, that the local solution method is an effective and flexible method. In particular, with the local solution method, we will give a unified proof of three of the classical formulas, namely Jacod's formula, the progressive enlargement formula and the enlargement formula with honest time. We also show that, besides its generality, this method, necessiting long (but interesting) computations, may not be optimal.
In the context of a general continuous financial market model, we study whether the additional information associated with an honest time gives rise to arbitrage profits. By relying on the theory of progressive enlargement of filtrations, we explicitly show that no kind of arbitrage profit can ever be realised strictly before an honest time, while classical arbitrage opportunities can be realised exactly at an honest time as well as after an honest time. Moreover, stronger arbitrages of the first kind can only be obtained by trading as soon as an honest time occurs. We carefully study the behavior of local martingale deflators and consider no-arbitrage-type conditions weaker than NFLVR.
Let $\mathbb{F}$ be a filtration and $τ$ be a random time. Let $\mathbb{G}$ be the progressive enlargement of $\mathbb{F}$ with $τ$. We study the validity of the following formula, called optional splitting formula : For any $\mathbb{G}$-optional process $Y$, there exist a $\mathbb{F}$-optional process $Y'$ and a function $Y"$ defined on $[0,\infty]\times(\mathbb{R}_+\timesΩ)$ being $\mathcal{B}[0,\infty]\otimes\mathcal{O}(\mathbb{F})$ measurable, such that $$ Y=Y'\ind_{[0,τ)}+Y"(τ)\ind_{[τ,\infty)} $$ We are interested in this formula, because it has been taken for granted in number of recent works in credit risk modeling, whilst such a formula can not be true in general. Sufficient conditions will be given for the validity of the above formula as well as of its extension in the case of multiple random times.
Given a stochastic structure with a filtration $\mathbb{F}$, the class of all random times whose conditional distribution functions are differentiable with respect to some $\mathbb{F}$ adapted non decreasing processes is considered. The main property of a random time in this class is that it can be isomorphically implanted into an auxiliary model which is absolutely continuous with respect to a Cox model. Three formulas are established: the conditional expectation formula, the optional splitting formula, and the enlargement of filtration formula. This study is particularly useful for models which are not defined directly with Cox models, such as the dynamic one-default model developed recently.
We consider a viable market model. Suppose that new information arrives at the market. We are interested in modeling the market reaction facing to the change of information. In particular we seek for the limit on the intensity of information change below which the market stays always viable. We succeed to find such a limit with the drift operator when the market possesses the martingale representation property.