Novel Esscher concepts for informational risks and valuation in market models under random horizon

summary

The gist

Novel Esscher concepts for informational risks and valuation in market models under random horizon considers how Esscher densities and measures can evolve with informational risk induced by a random…

In short

The research introduces novel Esscher pricing densities and measures to value claims under informational risks arising from a random time $ au$. This new concept extends existing methods by explicitly incorporating pure default risks, providing a rigorous framework for valuation in market models with stochastic volatility and jump-diffusion dynamics.

Key concepts

Esscher Pricing Densities/Measures
These are mathematical tools used to determine the fair price of financial claims. The novel concept extends previous ideas by adding a degree of freedom that allows the pricing measure to explicitly capture pure default risks, making valuations more precise in complex markets.
Informational Risk ($ au$)
This risk arises because market information is not perfectly known, and this uncertainty is modeled by a random time $ au$. The model accounts for how this informational risk evolves over time, influencing the pricing measures and claim valuation.
Geometric Jump-Diffusion Model
This describes how the underlying asset price (S) moves. It incorporates both continuous changes (like Brownian motion) and sudden, large price jumps, which are characteristic of real-world financial markets with stochastic volatility.

Terminology used across episodes

This episode discusses

The paper

Novel Esscher concepts for informational risks and valuation in market models under random horizon · Read on arXiv

Tahir Choulli, Ella Elazkany

University of Alberta

This paper considers the setting constituted by the triplet (S, F,τ). The pair (S, F) is the initial model where S is the discounted price process of risky assets (Stocks) and F is the ``public" flow of information that is available to all agents over time. τ is an arbitrary random time, which might not be seen through the flow F, and it represents a default time in credit risk and a death time of an insured in life insurance where mortality and/or longevity risks present serious challenges. For simplicity, we suppose S follows a geometric jump-diffusion model with stochastic volatiles and F is generated by a Brownian motion and Poisson process. In this setting, we address how the Esscher densities and/or measures notions can evolves with the informational risk induced by τ. In fact, we introduce the notion of Esscher with degrees-of-freedom associated to the informational risks generated by τ. On the one hand, we give various and complete characterizations of these Esscher notions in our informational setting. On the other hand, we fully characterize the Escher-prices bounds, for vulnerable claims (claims built on the occurrence of τ), using constrained backward stochastic differential equations (CBSDEs for short hereafter) that appear to be novel as well.

Transcript

Introduction to the show: ident: Quantitative Finance. The day in quantitative finance: models of markets, risk and prices.

Ruth: We're looking at the paper: "Novel Esscher concepts for informational risks and valuation in market models under random horizon".

Warren: Here's the gist of the paper. Novel Esscher concepts for informational risks and valuation in market models under random horizon considers how Esscher densities and measures can evolve with informational risk induced by a random time τ 5, five or more words copied word for word from that page.

Ruth: The model assumes S follows a geometric jump-diffusion model with stochastic volatiles and F is generated by a Brownian motion and Poisson process 3, five or more words copied word for word from that page.

Warren: A new flow of information G results from the progressive enlargement of F with τ, leading to the informational model (Sτ, G, P) 5, five or more words copied word for word from that page.

Ruth: Novel Esscher Concepts The authors introduce a novel family of Esscher pricing densities/measures for vulnerable claims taking into consideration the informational risks 3, five or more words copied word for word from that page.

Title and authors: Ruth: Let's talk about what the title of this paper, "Novel Esscher concepts for informational risks and valuation in market models under random horizon," actually signals. It’s pointing toward a new way to value things when you have uncertainty that doesn't just happen at a fixed point in time.

Warren: I think the word "novel" suggests they are extending what's already been done, specifically the Esscher transform, which is used for changing probability measures to get a positive linear pricing rule.

Iris: They are extending the second-order Esscher notion introduced recently in a prior paper, and this new idea allows them to explicitly capture pure default risks by adding an extra degree of freedom.

Ruth: That means they're not just looking at standard risk; they can now explicitly model that pure default risk component through this new structure.

Warren: The characterization involves a quadruplet, theta, psi, and phi, defining the Esscher density or measure as exp theta times S tau plus something involving X times delta S tau times psi delta S tau plus phi times N G minus gamma G squared.

Iris: That formula shows how they define this density or measure Z using that quadruplet to account for all these factors.

Ruth: So, what this means practically is that we can now price vulnerable claims in a way that specifically accounts for the impact of informational risk driven by tau.

Warren: It's about making the pricing measure more sensitive to how much new information, represented by G, is coming in as time progresses.

Iris: This shifts the focus from just standard asset volatility to incorporating these specific types of informational shocks into the valuation framework.

The paper's summary: Ruth: Moving on to what the core of this paper actually summarizes, they are tackling the evaluation problem for these vulnerable claims in this informational setting. They are focusing on using Esscher pricing as their primary method.

Warren: They explain that since there are many pricing methods out there, they zero in on using the Esscher pricing concept because it gives them a consistent positive linear rule when you change the probability measure.

Iris: They build on work by Gerber and Shiu to extend this transform to cases where the logarithm of the stock is a stochastic process with stationary and independent increments.

Ruth: The paper then discusses how the impact of that extension was significant in both actuarial science and finance, referencing citations like twenty-seven <ref:2611.16129#pg2>.

Warren: Then they introduce what they call the dynamical Esscher transform for discrete time models, which is another key milestone in evolving this concept.

Iris: The main result here is proving the convergence of associated quadruples of processes within the appropriate functional spaces, which leads to a rigorous existence result for the upper Esscher price process.

Ruth: So, they're not just proposing a new idea; they are providing a complete mathematical foundation for finding these pricing bounds even when default risk is present.

Warren: They show that by demonstrating that their approximating solutions converge to the solution of the constrained pricing backward stochastic differential equation, they get this rigorous existence result.

Iris: This essentially supplies the existence theory for the upper Esscher price process in a situation where default risk is involved, which was previously unavailable in the literature.

The paper's improvements: Ruth: Now let's look at what the authors suggest are the key improvements they made to this line of research. They focus on extending the existing theory to handle these informational risks more robustly.

Warren: They introduce a new family of Esscher pricing densities or measures specifically designed for this informational risk framework, building upon that second-order Esscher methodology we talked about earlier.

Iris: This new notion extends the second-order Esscher idea from a previous work and allows the pricing measures to explicitly capture pure default risks through this additional degree of freedom.

Ruth: That’s significant because it moves beyond just modeling volatility and starts incorporating the specific risk associated with default events directly into the measure itself.

Warren: The characterization involves that quadruplet we mentioned earlier, theta, psi, and phi defining the density or measure Z = exp theta times S tau plus X(delta S tau)tr psi delta S tau plus phi times N G minus gamma G squared.

Iris: They provide three equivalent characterizations of the Esscher density of order two with that degree of freedom phi, relating the G-Esscher density to the F-Esscher density and establishing necessary and sufficient conditions for their existence.

Ruth: This gives us a solid mathematical toolset, not just one idea, to characterize how these densities behave in this complex informational setting.

Warren: The paper also characterizes the bounds of Esscher price intervals using constrained backward stochastic differential equations or CBSDEs. They show the upper Esscher price process is the smallest solution to constrained linear BSDEs that appear new.

Iris: This provides a rigorous existence result for the upper Esscher price process specifically in the presence of default risk, which was previously lacking.

Conclusion: Ruth: So, to wrap up this discussion on "Novel Esscher concepts for informational risks and valuation in market models under random horizon," these authors have provided a complete mathematical foundation for pricing vulnerable claims under informational risk.

Warren: They’ve essentially given us the tools to move beyond simpler models and explicitly account for the impact of default times or other random events that aren't immediately visible in standard asset price movements.

Iris: The main implication is providing a complete characterization of pricing bounds, which was previously unavailable, by using these constrained backward stochastic differential equations.

Ruth: It’s a big step because it takes the Esscher transform and makes it work in this more complex, information-dependent environment where time itself is random.

Warren: The paper establishes that their approximating sequence converges to the upper bound process through martingale properties and Girsia's lemma, confirming that the upper bound is indeed the smallest solution to a constrained BSDE with a Skorokhod condition.

Iris: From an econometrician's view, this convergence proof gives us confidence in using these bounds for practical valuation problems involving default risk.

Ruth: So, we’ve seen how they use Esscher concepts to tackle informational risks under random horizons. It’s a deep dive into how market models can evolve when the flow of information itself is probabilistic and time-dependent.

Warren: It opens up new avenues for modeling claims where the timing of an event, like a default, is uncertain and tied to some external information process.

Iris: The work provides a rigorous framework, but it’s important to remember that the model assumes S follows a geometric jump-diffusion model with stochastic volatilities and F is generated by a Brownian motion and Poisson process.

Ruth: Exactly. That's the starting point they build everything on, so when you apply this concept, you have to be thinking about what those initial assumptions imply for your market participants.

Warren: It’s a lot of machinery to run, so running these models at scale will certainly require some serious computational resources.

More episodes

← Home