Paper: arXiv 2401.02601
Authors: Albert Dorador
Abstract
We propose an alternative linearization to the classical Markowitz quadratic portfolio optimization model, based on maximum drawdown. This model, which minimizes maximum portfolio drawdown, is particularly appealing during times of financial distress, like during the COVID-19 pandemic. In addition, we will present a Mixed-Integer Linear Programming variation of our new model that, based on our out-of-sample results and sensitivity analysis, delivers a more profitable and robust solution with a 200 times faster solving time compared to the standard Markowitz quadratic formulation.
Complexity vs Empirical Score
- Math Complexity: 7.0/10
- Empirical Rigor: 5.0/10
- Quadrant: Holy Grail — high math complexity, high empirical rigor
Why this score: The paper introduces a novel linearization of the Markowitz model using maximum drawdown, involving advanced optimization techniques like Mixed-Integer Linear Programming (MILP) and sensitivity analysis, which increases math complexity. However, while it reports out-of-sample results and performance metrics, the excerpt lacks detailed implementation code, specific datasets, or extensive statistical validation, keeping empirical rigor moderate.
Research Flowchart
flowchart TD A["Research Goal: Robust & Fast Portfolio Optimization"] --> B["Methodology: MILP Formulation<br>(Constrained Max Drawdown)"] B --> C["Data/Inputs: Historical Returns<br>(e.g., COVID-19 Period)"] C --> D["Computational Process: MILP Solver"] D --> E["Key Outcomes: <br>1. 200x Faster Solving Time <br>2. More Profitable & Robust Results"] E --> F["Comparison: Outperforms Standard Markowitz"]
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 linearized portfolio optimization model based on minimizing maximum drawdown, including a mixed‑integer linear programming version that solves much faster than the classic Markowitz quadratic formulation.
Source sentence
We propose an alternative linearization to the classical Markowitz quadratic portfolio optimization model, based on maximum drawdown.
Assumes 3
- Parameter values are correctly observed and fixed in deterministic portfolio optimization models. model
Source sentence
In all deterministic portfolio optimization models the general assumption is that parameter values are correctly observed (i.e. sample values correspond to true population values) and are fixed, at least to a relevant extent and for sufficiently long.
- The distribution of returns stays constant during the investment horizon or follows a known model. statistical
Source sentence
e.g. the distribution of returns stays constant during the investment horizon, or at most it changes according to a known model
- A 50‑50 train/test split is used, assuming the past three months of data represent future performance. data
Source sentence
We consider a 50-50 train/test split for our data. Concretely, we use February 1 – May 1, 2020 data to train and May 2 – August 1, 2020 to test.
Data
- Early COVID‑19‑era market data market-prices public
Source sentence
Therefore, we use early COVID-19-era market data on about 400 US stocks from the S& P 500 index, publicly available on Yahoo Finance and automatically downloaded through an in-house Python script.
Costs
Not modeled
Finds 1
- The MILP variation of the max‑drawdown model solves about 200 times faster than the quadratic Markowitz formulation. performance
Source sentence
our MILP variation is the fastest, by far: 200 times faster than the QPs, and 25 times faster than the unconstrained MD (LP).
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 8 related papers). Coverage is partial, so this is neither replication nor refutation.
2 papers in the archive
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.