Paper: arXiv 2407.00887
Authors: Cristiano Arbex Valle
Abstract
Portfolio optimisation is essential in quantitative investing, but its implementation faces several practical difficulties. One particular challenge is converting optimal portfolio weights into real-life trades in the presence of realistic features, such as transaction costs and integral lots. This is especially important in automated trading, where the entire process happens without human intervention. Several works in literature have extended portfolio optimisation models to account for these features. In this paper, we highlight and illustrate difficulties faced when employing the existing literature in a practical setting, such as computational intractability, numerical imprecision and modelling trade-offs. We then propose a two-stage framework as an alternative approach to address this issue. Its goal is to optimise portfolio weights in the first stage and to generate realistic trades in the second. Through extensive computational experiments, we show that our approach not only mitigates the difficulties discussed above but also can be successfully employed in a realistic scenario. By splitting the problem in two, we are able to incorporate new features without adding too much complexity to any single model. With this in mind we model two novel features that are critical to many investment strategies: first, we integrate two classes of assets, futures contracts and equities, into a single framework, with an example illustrating how this can help portfolio managers in enhancing investment strategies. Second, we account for borrowing costs in short positions, which have so far been neglected in literature but which significantly impact profits in long/short strategies. Even with these new features, our two-stage approach still effectively converts optimal portfolios into actionable trades.
Complexity vs Empirical Score
- Math Complexity: 6.5/10
- Empirical Rigor: 8.0/10
- Quadrant: Holy Grail — high math complexity, high empirical rigor
Why this score: The paper introduces a two-stage mathematical framework involving optimization formulations and goal programming, but it is grounded in solving practical computational constraints rather than abstract theory. It includes extensive computational experiments and addresses real-world implementation hurdles like transaction costs and asset integration, indicating strong empirical validation.
Research Flowchart
flowchart TD
A["Research Goal:<br>Bridge gap between<br>portfolio theory and practice"] --> B{"Two-Stage Framework"}
B --> C["Stage 1: Optimize<br>Portfolio Weights<br>(Target Allocation)"]
B --> D["Stage 2: Convert to<br>Realistic Trades<br>(Accounting for Costs)"]
C --> E{"Incorporate Novel Features"}
E --> F["Futures Contracts<br>& Equities Integration"]
E --> G["Borrowing Costs<br>for Short Positions"]
D --> H["Computational Results:<br>Mitigates intractability,<br>numerical imprecision,<br>and modeling trade-offs"]
F --> I["Key Outcome:<br>Actionable trading strategy<br>enhanced with realistic features"]
G --> I
H --> I
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 two‑stage optimisation framework that first determines optimal portfolio weights and then generates realistic trades.
Source sentence
propose a two-stage framework as an alternative approach to address this issue.
Assumes 6
- Trading is executed at the end of the day using closing prices. execution
Source sentence
We assume trading at the end of the day based on closing prices, 5 basis points as variable transaction costs and no fixed costs.
- Variable transaction costs are 5 basis points per trade and there are no fixed costs. execution
Source sentence
We assume trading at the end of the day based on closing prices, 5 basis points as variable transaction costs and no fixed costs.
- Stocks are traded in round lots of 100 shares or in odd lots (any integral number of shares). market
Source sentence
We assume that S&P500 stocks are either traded in round lots (consisting of 100 shares each) or in odd lots (any integral number of shares).
- An asset is eligible for rebalancing only if at most 5% of its recent price history is missing. data
Source sentence
An asset is only considered eligible in a rebalance if there are at most 5% prices missing from the most recent historical series of 201 prices.
Show 2 more
- Cash earns the 10‑year US Treasury bond return and is used as a synthetic asset for lent/borrowed cash. model
Source sentence
Interest in cash is accrued every day, according to the 10-year US treasury bond (symbol TNX) returns. This series is used as a synthetic asset representing the return on both lent and borrowed cash.
- Borrowing costs are applied to short positions in long/short strategies. execution
Source sentence
Borrowing costs are applicable in the long/short strategies, which further impact the strategy performance.
Data
2012-12 – 2023-12 · S&P500 stocks and futures · multi-asset · monthly
- S&P500 stock dataset market-prices unknown
Source sentence
... out-of-sample period spans from 25th December 2012 up to 29th December 2023.
- Expanded dataset with stocks and futures market-prices unknown
Source sentence
We use here the expanded dataset described in Section 4.1, with stocks and futures.
Costs
Modeled
5 basis points per trade (variable) and no fixed costs; borrowing costs also applied to short positions.
Source sentence
We assume trading at the end of the day based on closing prices, 5 basis points as variable transaction costs and no fixed costs.
Finds 4
Sharpe 1.00 (benchmark 0.59) · annual return 19.9% · max drawdown 24.2%
- The two‑stage approach mitigates computational intractability and numerical imprecision of single‑stage models while remaining tractable for reasonably sized portfolios. theoretical
Source sentence
Through extensive computational experiments, we show that our approach not only mitigates the difficulties discussed above but also can be successfully employed in a realistic scenario.
- Including odd‑lot trading and transaction costs reduces CAGR by about 0.73 % and Sharpe ratio by ~4 % relative to the cost‑free case. performance
Source sentence
the addition of odd lots and costs resulted in a decrease of 0.73% in CAGR, a ≈4% decrease in Sharpe and ≈3% in Sortino ratios.
- Borrowing costs in long/short strategies further lower performance, decreasing CAGR by 2.77 % (odd lots) and 2.94 % (round lots). performance
Source sentence
With odd lots we observe a decrease of 2.77% in CAGR, 11% in the Sharpe and Sortino ratios, while with round lots we observe drops of 2.94% in CAGR, 11% in the Sharpe ratio and 12% in the Sortino ratio.
- When the second stage does not fail, its solutions are only marginally worse than the exact single‑stage solutions, and the gap shrinks as portfolio size grows. theoretical
Source sentence
comparisons with an existing method show that, when it does not fail, the benefits of a single stage over a two-stage approach are somewhat marginal.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 19 related papers). Coverage is partial, so this is neither replication nor refutation.
1 paper in the archive
Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.