Paper: arXiv 2309.05926

Authors: Igor Halperin

Abstract

We consider the problem of optimization of contributions of a financial planner such as a working individual towards a financial goal such as retirement. The objective of the planner is to find an optimal and feasible schedule of periodic installments to an investment portfolio set up towards the goal. Because portfolio returns are random, the practical version of the problem amounts to finding an optimal contribution scheme such that the goal is satisfied at a given confidence level. This paper suggests a semi-analytical approach to a continuous-time version of this problem based on a controlled backward Kolmogorov equation (BKE) which describes the tail probability of the terminal wealth given a contribution policy. The controlled BKE is solved semi-analytically by reducing it to a controlled Schrodinger equation and solving the latter using an algebraic method. Numerically, our approach amounts to finding semi-analytical solutions simultaneously for all values of control parameters on a small grid, and then using the standard two-dimensional spline interpolation to simultaneously represent all satisficing solutions of the original plan optimization problem. Rather than being a point in the space of control variables, satisficing solutions form continuous contour lines (efficient frontiers) in this space.

Complexity vs Empirical Score

  • Math Complexity: 8.5/10
  • Empirical Rigor: 2.0/10
  • Quadrant: Lab Rats — high math complexity, limited empirical testing

Why this score: The paper involves advanced mathematics including stochastic differential equations, backward Kolmogorov equations, reduction to Schrödinger equations, and semi-analytical solutions, placing it at high math complexity. However, the empirical rigor is low as the approach relies on theoretical derivations and numerical interpolation on a small grid without evidence of backtesting on real data, coding implementations, or statistical performance metrics.

Research Flowchart

  flowchart TD
  A["Research Goal: Optimal contribution plan<br>to meet financial goal at a given confidence level"] --> B{"Key Methodology"}
  B --> C["Semi-analytical approach using<br>Controlled Backward Kolmogorov Equation BKE"]
  C --> D["Reduce BKE to<br>Controlled Schrödinger Equation"]
  D --> E["Solve Schrödinger Eq<br>via algebraic method"]
  E --> F{"Computational Process"}
  F --> G["Compute solutions on fine grid<br>for all control parameters"]
  G --> H["Interpolate solutions using<br>2D Spline for continuous representation"]
  H --> I["Key Findings/Outcomes"]
  I --> J["Efficient Frontiers: Satisficing solutions<br>form continuous contours in control space"]
  I --> K["Objective: Find optimal & feasible<br>contribution schedules for goal-based planning"]

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 semi‑analytical approach to the continuous‑time optimal contribution problem by solving a controlled backward Kolmogorov equation via reduction to a controlled Schrödinger equation and using algebraic methods.

Source sentence
This paper suggests a semi-analytical approach to a continuous-time version of this problem based on a controlled backward Kolmogorov equation (BKE) which describes the tail probability of the terminal wealth given a contribution policy. The controlled BKE is solved semi-analytically by reducing it to a controlled Schrodinger equation and solving the latter using an algebraic method.

Assumes 6

  • The portfolio maintains a fixed equity‑to‑total ratio ω (e.g., ω = 0.9). model
    Source sentence
    expressed as a desired terminal wealth at time T. The investment portfolio has a fixed ratio ω of equity to the total portfolio value (e.g. ω = 0.9), which is assumed to be maintained by a fiduciary that manages the portfolio on client’s behalf.
  • Transaction costs are proportional to the amount of equity purchased: ν ω |c_t| Δt. execution
    Source sentence
    This investment of cash from the planner forces the fiduciary to buy the amount of equity equal to ωct∆t, which also incurs proportional transaction cost νω|ct|∆t, where ν is a parameter.
  • Contributions are non‑negative (c_t ≥ 0), allowing omission of the absolute value in the cost formula. other
    Source sentence
    As we have ct ≥0 in our problem, in what follows we omit the absolute value expression in the transaction cost formula.
  • Equity returns are normally distributed with mean ˆr_e and volatility ˆσ_e (Gaussian shocks ξ_t ∼ N(0,1)). market
    Source sentence
    The cash and equity positions at time t+ are then obtained as follows: b+ t = bt + ((1 −ω)ct −νωct) ∆t s+ t = st + ωct∆t (1) This produces the following update rule for the total portfolio value Πt := bt + st: Π+ t = Πt + (ct −νωct) ∆t (2) We assume that equity has random returns re(t) = ˆre + ˆσe √ ∆t ξt, ξt ∼N(0, 1) (3) with a mean value ˆre, while the cash component has a deterministic return rf.
Show 2 more
  • Contribution policies are parameterized by an initial rate u₀ and exponential growth rate ξ: u_t = u₀ e^{ξt}. model
    Source sentence
    we consider contribution policies parameterized in terms of an initial contribution rate u0 and a growth rate ξ: ut = u0eξt (6) so that u0 and ξ are now the constrained decision variables for the planner.
  • The controlled SDE can be transformed to a dimensionless variable v_t that embeds u₀ in the initial condition, leaving ξ as the only explicit control parameter. model
    Source sentence
    2.2 Stochastic Verhulst equation and Langevin dynamics Note that Eq.(4) is a controlled SDE where control variables are u0 and ξ. To simplify these dynamics, we introduce a new dimensionless variable vt defined as follows: vt = 2 σ2 ut Πt = 2 σ2 u0eξt Πt , (7) Note that the new variable vt blends together the state variable Πt and action variable ut.

Data

  • Synthetic retirement plan example synthetic unknown
    Source sentence
    In Fig. 2, we show 3D views of the WF ¯Ψ(y, T) and the BKE density p(y, 0) as a function of two variables y = y0 and ξ for a typical retirement plan with the horizon T = 20Y for the target wealth of $2.5m, initial wealth of $0.5m, and portfolio volatility of 30%, with the following bounds on admissible values of u0 and ξ: $10k ≤u0 ≤$100K, and 2.5% ≤ξ ≤5.0%.

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 (Proportional cost ν ω |c_t| Δt is included in the portfolio dynamics and in the update rule for total portfolio value.); the classifier found no reported trading result, so costs are not applicable.

Finds 3

  • Efficient frontier lines (constant‑probability contours) can be visualized as continuous contour lines in the control‑parameter space, enabling real‑time policy optimization and scenario analysis. theoretical
    Source sentence
    Efficient frontier lines shown in Figs. 3 and 4 can be used for visualization of all possible optimal policies at different confidence levels α for a given plan with user-defined initial and target wealth levels and constraints on admissible values of control parameters. Note that while our original objective was formulated as an optimization problem, the solution presented here reduces to a combination of a 2D
  • The plan optimization reduces to a 2‑D cubic spline interpolation combined with a root‑finding (contour construction) algorithm, eliminating the need for iterative optimization. theoretical
    Source sentence
    Note that while our original objective was formulated as an optimization problem, the solution presented here reduces to a combination of a 2D cubic spline and a root-finding, or equivalently a contour construction algorithm.10 The plan optimization part thus reduces to the task of drawing a constant-probability line on a computed solution surface given by the BKE density p(y, t|ξ). Figure 3: 3D plot of the BKE tail
  • The second transformation of the BKE to a Schrödinger equation (Eq. 17) is novel in the stochastic control literature. theoretical
    Source sentence
    To our knowledge, the second transformation in Eq.(17) for the BKE equation was not previously considered in the stochastic control literature.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 9 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.