Paper: arXiv 2412.12458

Authors: Jirat Suchato, Sean Wiryadi, Danran Chen, Ava Zhao, Michael Yue

Abstract

We conduct a preliminary analysis of a pairs trading strategy using the Ornstein-Uhlenbeck (OU) process to model stock price spreads. We compare this approach to a naive pairs trading strategy that uses a rolling window to calculate mean and standard deviation parameters. Our findings suggest that the OU model captures signals and trends effectively but underperforms the naive model on a risk-return basis, likely due to non-stationary pairs and parameter tuning limitations.

Complexity vs Empirical Score

  • Math Complexity: 6.5/10
  • Empirical Rigor: 4.0/10
  • Quadrant: Lab Rats — high math complexity, limited empirical testing

Why this score: The paper employs advanced stochastic calculus and econometric testing (cointegration, OU process SDEs), but the empirical validation is preliminary, using a limited dataset without robust out-of-sample testing or risk management refinements.

Research Flowchart

  flowchart TD
  A["Research Goal: Compare OU Process Model vs. Naive Rolling Window<br>for Pairs Trading Signals"] --> B
  subgraph B ["Methodology & Inputs"]
      direction LR
      B1["Pair Selection<br>Stock Price Data"] --> B2["Spread Calculation<br>Price Ratio or Log Spread"]
  end
  B --> C{"Computational Modeling"}
  C --> D["Naive Strategy<br>Rolling Mean & STD"]
  C --> E["OU Process Strategy<br>MLE Parameter Estimation"]
  D --> F
  E --> F["Strategy Evaluation<br>Risk-Return Metrics"]
  F --> G["Findings & Outcomes"]
  subgraph G
      direction TB
      G1["OU Model captures<br>signals & trends effectively"]
      G2["Naive Model performs better<br>on risk-return basis"]
      G3["Limitations: Non-stationary pairs,<br>parameter tuning challenges"]
  end

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 pairs‑trading strategy that models stock price spreads with an Ornstein‑Uhlenbeck process and compares it to a naive rolling‑window approach.

Source sentence
we employ the Ornstein-Uhlenbeck process to do so, and we compare our results to a basic computation of the z score using the rolling mean and standard deviation by backtesting our results on historical pricing data.

Assumes 5

  • Trades are executed the next day with negligible size that does not move the market. execution
    Source sentence
    We assume that we transact on the next day and that the size of trades is negligible and does not move the market.
  • All selected pairs exhibit mean‑reverting behavior. model
    Source sentence
    The strategy also relies on the assumption that all the selected pairs exhibits mean-reverting behavior when in reality this assumption may not hold.
  • Transaction costs and slippage are ignored in the simulation. execution
    Source sentence
    The simulation did not incorporate transaction costs and if this strategy were to be implemented in real life the returns could be even lower. Slippage was also not accounted for.
  • Pairs are stationary; non‑stationarity may degrade performance. statistical
    Source sentence
    We suspect that this is likely due to potential non-stationarity of some pairs which led to poor mean reversion.
Show 1 more
  • Data are limited to a one‑year time frame starting in 2018. data
    Source sentence
    The strategy was tested on a limited one-year time frame, which may not fully capture the effects of varying economic conditions...

Data

Backtest sample: equities · daily

  • Refinitiv Datastream wrds ds2dsf table · Refinitiv market-prices commercial
    starting from 2018 · US‑based companies with market cap > $1 billion
    Source sentence
    We limited the dataset to companies listed in the US, trading in USD, and filtered for data starting from 2018.

Recognised sources: Refinitiv / LSEG (Datastream, Eikon, Tick History)

Costs

Not modeled A trading or execution result is reported without costs.

Source sentence
The simulation did not incorporate transaction costs and if this strategy were to be implemented in real life the returns could be even lower. Slippage was also not accounted for.

Finds 4

Reported performance (Ornstein-Uhlenbeck pairs trading): Sharpe 0.47 (benchmark 0.78) · annual return 4.9% · max drawdown 9.0%

  • The OU‑based pairs‑trading portfolio achieved an annualized expected return of 4.92% with a Sharpe ratio of 0.468, underperforming the naive strategy. performance
    Source sentence
    We find that the portfolio using the Ornstein-Uhlenbeck process has an expected return of 4.92% with a Sharpe ratio of 0.468.
  • The naive (standard) pairs‑trading portfolio achieved an annualized expected return of 7.21% with a Sharpe ratio of 0.778. performance
    Source sentence
    We find that the standard portfolio has an expected return of 7.21% with a Sharpe ratio of 0.778.
  • The OU strategy produced a positive daily return 53.3% of the time, slightly lower than the standard strategy’s 54.63% win‑ratio. performance
    Source sentence
    Our findings show that the standard pairs trading strategy delivered a positive daily return 54.63 % of the time and the OU pairs trading strategy deliver a positive daily return 53.3 % of the time.
  • Despite lower risk‑adjusted performance, the OU model captured correct signals and overall trend, indicating strong predictive power but suffering from suboptimal parameter tuning. negative
    Source sentence
    Although the OU Model performed worse than the standard model, the cumulative returns plot indicates that the model captures the correct signals and overall trend

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 39 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.