Linear Approximation of Market Impact for Portfolio Optimization

Summary

The document summarizes a simplified market impact model for large investment portfolios. It builds on an Almgren-style impact function whose nonlinear power form can make portfolio optimization slow and can leave uncertainty about whether the numerical solution is globally optimal. The proposed approach approximates that function with piecewise linear segments, allowing the optimization problem to be expressed as a quadratic program.

The reported empirical comparison found portfolio results from the simplified model to be very close to those from the original model, while the reformulation makes solving the optimization problem faster and guarantees a global optimum for the stated numerical problem. The author argues that this practical approximation is appropriate because market impact is not directly observable and need not be modeled with extreme precision. The summary gives no datasets, detailed calibration procedure, or numerical performance measures. It flags model failure and parameter estimation error as risks, so its conclusions depend on the suitability of the assumed impact function and its estimated parameters.

Key ideas

• The model approximates a nonlinear market impact function with piecewise linear segments.
• The approximation recasts portfolio optimization as a quadratic programming problem.
• The report says the simplified and original models produced very similar portfolio performance.
• The reformulated problem is described as faster to solve with a globally optimal numerical solution.
• Market impact estimates remain uncertain, and model failure or parameter error can undermine results.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.