Paper: arXiv 2310.08284
Authors: Fredi Šarić, Stjepan Begušić, Andro Merćep, Zvonko Kostanjčar
Abstract
Statistical arbitrage methods identify mispricings in securities with the goal of building portfolios which are weakly correlated with the market. In pairs trading, an arbitrage opportunity is identified by observing relative price movements between a pair of two securities. By simultaneously observing multiple pairs, one can exploit different arbitrage opportunities and increase the performance of such methods. However, the use of a large number of pairs is difficult due to the increased probability of contradictory trade signals among different pairs. In this paper, we propose a novel portfolio construction method based on preference relation graphs, which can reconcile contradictory pairs trading signals across multiple security pairs. The proposed approach enables joint exploitation of arbitrage opportunities among a large number of securities. Experimental results using three decades of historical returns of roughly 500 stocks from the S&P 500 index show that the portfolios based on preference relations exhibit robust returns even with high transaction costs, and that their performance improves with the number of securities considered.
Complexity vs Empirical Score
- Math Complexity: 7.5/10
- Empirical Rigor: 8.5/10
- Quadrant: Holy Grail — high math complexity, high empirical rigor
Why this score: The paper introduces complex graph theory and preference relation frameworks (e.g., potential method) requiring advanced mathematics for signal reconciliation, while providing extensive empirical validation with 30 years of S&P 500 data and robustness testing against transaction costs.
Research Flowchart
flowchart TD A["Research Goal:<br>Reconcile Contradictory<br>Pairs Trading Signals"] --> B["Methodology:<br>Preference Relation Graphs"] B --> C["Data Input:<br>Historical Returns<br>~500 S&P 500 Stocks<br>30 Years"] C --> D["Process:<br>Construct Preference Graph<br>and Reconcile Signals"] D --> E["Output:<br>Robust Portfolio<br>Construction"] E --> F["Key Outcome 1:<br>Sustained Returns<br>Even w/ High Transaction Costs"] E --> G["Key Outcome 2:<br>Performance Improves<br>with More Securities"]
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 novel portfolio construction method based on preference relation graphs that reconciles contradictory pairs trading signals.
Source sentence
we propose a novel portfolio construction method based on preference relation graphs
Assumes 5
- Portfolios are reweighted and rebalanced on a daily basis using open prices of the next day. execution
Source sentence
Since the portfolio formed by the proposed method is reweighted and rebalanced on a daily basis, the effect of slippage and transaction cost is significant, especially when portfolio turnover is high.
- The entire capital is assumed to be fully invested in the portfolio each day. execution
Source sentence
To calculate total returns over multiple time periods, the entire capital is assumed to be fully invested in the given portfolio each day, with the weights as defined in Eq. (13).
- Preference function outputs are standardized with mean 0 and standard deviation 1. model
Source sentence
A preference function whose outputs are standardized with mean 0 and standard deviation of 1 greatly simplifies the selection of thresholding parameter κ.
- Preference preserving transformations (thresholding, vertex pruning, final vertex selection) do not alter the underlying ranking. model
Source sentence
The removal of intermediate vertices in a preference graph is also a preference preserving transformation, as shown in appendix section B.
Show 1 more
- Bootstrap samples of securities are drawn without replacement to assess robustness. data
Source sentence
we use B = 100 security bootstrap samples of sizes N ′′ = 50 and N ′ = 250, and finally we evaluate strategies on the full set of N = 553 stocks.
Data
- Historical returns of U.S. stocks market-prices unknown
Source sentence
Experimental results using three decades of historical returns of roughly 500 stocks from the S&P 500 index
Costs
Modeled
0.1% per trade, higher than the usual 0.05% reported in pairs trading literature
Source sentence
... we use higher than usual transaction costs (most of the pairs trading research report a transaction costs of 0.05% (Avellaneda & Lee, 2010))
Finds 3
- The momentum decorator extends average holding periods up to 23 days and reduces daily turnover across all portfolio sizes. performance
Source sentence
... the momentum decorator drastically reduces turnover in all examined scenarios
- Using the momentum decorator lowers portfolio volatility (standard deviation) for all portfolio sizes. performance
Source sentence
the momentum decorator reduces the portfolio risk as measured by standard deviation, in all considered cases
- Utility proportional weighting yields similar mean returns to equal weighting but further reduces turnover when combined with the momentum decorator. performance
Source sentence
the inclusion of the momentum decorator improves the performance in terms of mean returns for the first two subsample sizes
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 12 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.