Paper: arXiv 2409.11569

Authors: Christoph Knochenhauer, Alexander Merkel, Yufei Zhang

Abstract

We consider the Merton problem of optimizing expected power utility of terminal wealth in the case of an unobservable Markov-modulated drift. What makes the model special is that the agent is allowed to purchase costly expert opinions of varying quality on the current state of the drift, leading to a mixed stochastic control problem with regular and impulse controls involving random consequences. Using ideas from filtering theory, we first embed the original problem with unobservable drift into a full information problem on a larger state space. The value function of the full information problem is characterized as the unique viscosity solution of the dynamic programming PDE. This characterization is achieved by a new variant of the stochastic Perron’s method, which additionally allows us to show that, in between purchases of expert opinions, the problem reduces to an exit time control problem which is known to admit an optimal feedback control. Under the assumption of sufficient regularity of this feedback map, we are able to construct optimal trading and expert opinion strategies.

Complexity vs Empirical Score

  • Math Complexity: 9.0/10
  • Empirical Rigor: 1.0/10
  • Quadrant: Lab Rats — high math complexity, limited empirical testing

Why this score: The paper is highly theoretical, employing advanced stochastic control, filtering theory, viscosity solutions, and the stochastic Perron method, while it lacks any empirical data, backtests, or implementation details, focusing entirely on mathematical proofs and existence results.

Research Flowchart

  flowchart TD
  A["Research Goal:<br>Optimal Investment with Costly Expert Opinions"] --> B["Methodology:<br>Stochastic Control & Filtering Theory"]
  B --> C["State Space Transformation:<br>Embed into Full Information Problem"]
  C --> D["Computational Process:<br>Solve DP PDE using<br>Stochastic Perron's Method"]
  D --> E["Control Reduction:<br>Identify Exit Time<br>Control Problem"]
  E --> F["Key Findings:<br>Characterize Value Function<br>Construct Optimal Strategies"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes a mixed stochastic control framework for optimal investment with costly expert opinions, characterizing the value function via a unique viscosity solution of a dynamic programming PDE and constructing optimal trading and expert opinion strategies under regularity assumptions.

Source sentence
This characterization is achieved by a new variant of the stochastic Perron's method, which additionally allows us to show that, in between purchases of expert opinions, the problem reduces to an exit time control problem which is known to admit an optimal feedback control. Under the assumption of sufficient regularity of this feedback map, we are able to construct optimal trading and expert opinion strategies.

Assumes 6

  • Expert opinions must be purchased for a strictly positive fee. execution
    Source sentence
    In our model, on the other hand, we assume that the agent has to purchase expert opinions in exchange for a strictly positive fee. As we will assume the cost for an expert opinion to be strictly positive, the agent’s expert opinion strategies are naturally modeled as impulse controls. 3 Definition 2.1.
  • The drift of the risky asset is modulated by a finite‑state continuous‑time Markov chain (Wonham setting). model
    Source sentence
    To be precise, we consider a financial market model similar to [32] in which there is a single risky asset with drift modulated by a finite-state continuous-time Markov chain, i.e. we place ourselves in the Wonham setting. We assume that Y takes values in {e1, . . . , eN}, the canonical unit vectors in RN, where N ∈N denotes the number of regimes of the economy.
  • The risk‑free asset has zero return and there is a single risky asset with constant volatility σ>0. market
    Source sentence
    The financial market is assumed to consist of a risk-free asset with zero returns and a single risky asset S = (Ss)s∈[0,T ] which satisfies dSs = Ss  µ⊺Ysds + σdBs  , S0 = s0, s ∈[0, T ], where s0 > 0 is the deterministic initial price, µ ∈RN are the returns in the respective states of the economy, and σ > 0 is the volatility.
  • The Brownian motion driving the risky asset is independent of the Markov chain. model
    Source sentence
    Finally, we take as given a one-dimensional standard (F, P)-Brownian motion B independent of Y .
Show 2 more
  • The underlying stochastic process satisfies the strong Markov property (or a martingale property sufficient for the impulse verification). model
    Source sentence
    One central assumption in this type of construction is that the underlying stochastic process satisfies the strong Markov property. This martingale property is sufficient to replace the strong Markov arguments in the impulse verification machinery of [3, 4, 5, 7].
  • The stochastic differential equation associated with the optimal control admits a strong solution. execution
    Source sentence
    We work around this issue by assuming that the stochastic differential equation associated with the optimal control admits a strong solution, hence allowing us to construct the optimal state process on arbitrary probability spaces and arbitrary driving Brownian motions.

Data

No dataset named in the text.

Costs

Not applicable The paper reports no trading or execution result; the cost the extractor found is a model parameter.

The paper-text extraction found a cost parameter inside a model (Expert opinion purchases incur a strictly positive fee; the fee is treated as an impulse cost in the control problem.); the classifier found no reported trading result, so costs are not applicable.

Finds 3

  • The value function of the full‑information problem is the unique viscosity solution of the associated HJB quasi‑variational inequality. theoretical
    Source sentence
    The value function of the full information problem is characterized as the unique viscosity solution of the dynamic programming PDE.
  • Between purchases of expert opinions the problem reduces to an exit‑time control problem that admits an optimal feedback control. theoretical
    Source sentence
    This characterization is achieved by a new variant of the stochastic Perron's method, which additionally allows us to show that, in between purchases of expert opinions, the problem reduces to an exit time control problem which is known to admit an optimal feedback control.
  • Under sufficient regularity of the feedback map, optimal trading and expert‑opinion strategies can be explicitly constructed. theoretical
    Source sentence
    Under the assumption of sufficient regularity of this feedback map, we are able to construct optimal trading and expert opinion strategies.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 18 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.