[Paper Review] Homogenization and asymptotics for small transaction costs: the multidimensional case
This paper establishes an asymptotic expansion for the multidimensional infinite-horizon optimal consumption-investment problem with small proportional transaction costs, using viscosity solutions and homogenization techniques. It proves the existence of a corrector solution to a singular ergodic control problem and derives the first-order correction term, which is characterized via a system of corrector equations without explicit solutions in higher dimensions.
In the context of the multi-dimensional infinite horizon optimal consumption-investment problem with proportional transaction costs, we provide the first order expansion in small transact costs. Similar to the one-dimensional derivation in our accompanying paper [42], the asymptotic expansion is expressed in terms of a singular ergodic control problem, and our arguments are based on the theory of viscosity solutions, and the techniques of homogenization which leads to a system of corrector equations. In contrast with the one-dimensional case, no explicit solution of the first corrector equation is available anymore. Finally, we provide some numerical results which illustrate the structure of the first order optimal controls.
Motivation & Objective
- To extend the one-dimensional asymptotic analysis of small transaction costs to the multidimensional setting.
- To characterize the first-order correction term in the expansion of the value function under small transaction costs.
- To establish the existence and regularity of a corrector solution to a singular ergodic control problem in high dimensions.
- To develop a homogenization framework where the fast variable emerges from the limit problem, not from initial oscillations.
- To provide numerical evidence for the structure of optimal trading strategies in the multidimensional case.
Proposed method
- Applies the viscosity solution approach of Evans to homogenization, combined with relaxed limits from Barles and Perthame.
- Uses a change of variables that introduces a fast variable dependent on the limit problem's optimal strategy, rather than predefining oscillatory terms.
- Derives a system of corrector equations linked to a singular ergodic control problem, analogous to the monotone follower problem.
- Employs recent results by Hynd (2013, 2014) to establish existence and $C^{1,1}$ regularity of the solution to the eigenvalue problem.
- Applies comparison principles and inf-convolution techniques to analyze the free boundary and second-order regularity.
- Uses uniform $C^{1,1}$ estimates and compactness arguments to pass to the limit and construct the corrector in the asymptotic regime.
Experimental results
Research questions
- RQ1How can the asymptotic expansion for the multidimensional optimal consumption-investment problem with small transaction costs be rigorously derived?
- RQ2What is the structure of the first-order correction term in the presence of multiple risky assets and proportional transaction costs?
- RQ3Can a corrector solution be constructed for the singular ergodic control problem that arises in the multidimensional setting?
- RQ4How does the homogenization framework adapt when the fast variable is not pre-specified but emerges from the limit problem?
- RQ5What are the regularity properties of the value function and its second-order derivatives in the asymptotic regime?
Key findings
- The first-order correction term in the asymptotic expansion is characterized by a singular ergodic control problem that arises from the homogenization process.
- A $C^{1,1}$ unique solution to the corrector equation exists and satisfies a precise growth condition, even without explicit solutions in higher dimensions.
- The second derivative of the corrector function is bounded almost everywhere and vanishes almost everywhere in the complement of the effective region.
- The value function's second-order regularity is preserved uniformly in the small-cost limit, enabling convergence in $C^1_{ ext{loc}}$.
- Numerical results illustrate that the first-order optimal controls exhibit a piecewise linear structure, consistent with the theoretical framework.
- The homogenization framework successfully handles non-periodic, non-oscillatory fast variables that emerge from the limit problem's structure.
Better researchstarts right now
From reading papers to final review, dramatically reduce your research time.
No credit card · Free plan available
This review was created by AI and reviewed by human editors.