Paper: arXiv 2512.17225

Authors: Dimitrios Bachtis, David S. Berman, Arabella Schelpe

Abstract

We use a \(φ^{“4”}\) quantum field theory with inhomogeneous couplings and explicit symmetry-breaking to model an ensemble of financial time series from the S\(&\)P 500 index. The continuum nature of the \(φ^4\) theory avoids the inaccuracies that occur in Ising-based models which require a discretization of the time series. We demonstrate this using the example of the 2008 global financial crisis. The \(φ^{“4”}\) quantum field theory is expressive enough to reproduce the higher-order statistics such as the market kurtosis, which can serve as an indicator of possible market shocks. Accurate reproduction of high kurtosis is absent in binarized models. Therefore Ising models, despite being widely employed in econophysics, are incapable of fully representing empirical financial data, a limitation not present in the generalization of the \(φ^{“4”}\) scalar field theory. We then investigate the scaling properties of the \(φ^{“4”}\) machine learning algorithm and extract exponents which govern the behavior of the learned couplings (or weights and biases in ML language) in relation to the number of stocks in the model. Finally, we use our model to forecast the price changes of the AAPL, MSFT, and NVDA stocks. We conclude by discussing how the \(φ^{“4”}\) scalar field theory could be used to build investment strategies and the possible intuitions that the QFT operations of dimensional compactification and renormalization can provide for financial modelling.

Complexity vs Empirical Score

  • Math Complexity: 9.0/10
  • Empirical Rigor: 3.0/10
  • Quadrant: Lab Rats — high math complexity, limited empirical testing

Why this score: The paper employs advanced quantum field theory concepts (ϕ⁴ theory, symmetry breaking, renormalization) and dense mathematical formulations, placing it in the high-complexity range, while its empirical validation is limited to a modest subset of stocks (20) and theoretical forecasting without live trading or extensive backtesting, resulting in low rigor.

Research Flowchart

  flowchart TD
  A["Research Goal:<br>Model S&P 500 financial time series<br>& forecast stock prices"] --> B["Methodology:<br>&phi;^4 Quantum Field Theory<br>(Inhomogeneous couplings, symmetry breaking)"]
  B --> C["Data Input:<br>S&P 500 Index & AAPL, MSFT, NVDA stocks"]
  C --> D["Computational Process:<br>Train &uuml;lf;^4 ML Algorithm<br>(Extract couplings & scaling exponents)"]
  D --> E["Outcome 1:<br>Reproduces high kurtosis<br>(Captures 2008 crisis shocks better than Ising)"]
  D --> F["Outcome 2:<br>Forecast price changes<br>(Investment strategy foundation)"]
  E & F --> G["Conclusion:<br>&phi;^4 theory generalizes better for financial data<br>Insights via dimensional compactification & renormalization"]

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 disordered φ⁴ quantum field theory, equivalent to a Markov random field machine learning algorithm, to model and forecast financial time series.

Source sentence
In this manuscript we propose the ϕ4 quantum field theory, a system with Z2 symmetry and continuous degrees of freedom, to model and forecast empirical financial time series.

Assumes 6

  • The subset of selected stocks forms a market. market
    Source sentence
    We assume that the subset of stocks forms a market and calculate statistics, such as the market mean and kurtosis.
  • Log‑returns are used as the time‑series representation. data
    Source sentence
    we consider the log-returns for the time series of a subset of V = 20 stocks from the S&P 500 index.
  • A simple moving average of 250 trading days smooths short‑term fluctuations. data
    Source sentence
    To smooth out short-term fluctuations, we apply a simple moving average of 250 days to the data, which corresponds to one trading year.
  • Training data precede prediction data chronologically. execution
    Source sentence
    The training set comprises the chronologically former values and the prediction set the most recent ones.
Show 2 more
  • Random subsets of stocks are used for finite‑size scaling analysis. statistical
    Source sentence
    We select V = 64 stocks from the S&P 500 index and calculate the average of their returns over 20 years at a fixed point in time.
  • Scaling exponents are obtained by numerical fits on logarithmic axes. statistical
    Source sentence
    The results of the finite size scaling analysis are depicted on logarithmic axes in Fig. 3.

Data

  • Subset of 20 S&P 500 stocks (log‑returns) · S&P 500 index market-prices public
    01/01/2013 onward for 2000 trading days · 20 S&P 500 constituents
    Source sentence
    The initial date for the time series is 01/01/2013 and we extend the time series for 2000 trading days in the past.
  • Subset of 64 S&P 500 stocks (log‑returns) · S&P 500 index market-prices public
    20‑year window ending 01/01/2023 · 64 S&P 500 constituents
    Source sentence
    We select V = 64 stocks from the S&P 500 index and calculate the average of their returns over 20 years at a fixed point in time: 01/01/2023.
  • AAPL, MSFT, NVDA daily returns market-prices public
    Three large‑cap US stocks
    Source sentence
    We train the model on a dataset comprising the returns for the time series of the AAPL, MSFT and NVDA stocks.

Costs

Not applicable The paper reports no trading or execution result.

Finds 5

  • The φ⁴ quantum field theory can reproduce higher‑order statistics such as market kurtosis, which Ising‑based binarized models cannot. statistical
    Source sentence
    The ϕ4 quantum field theory is expressive enough to reproduce the higher-order statistics such as the market kurtosis, which can serve as an indicator of possible market shocks.
  • Finite‑size scaling yields exponents kwij = –0.96(1) and kai = –0.81(1) (and similar values for a 2013 date). theoretical
    Source sentence
    through numerical fits we extract the values of two exponents kwij = −0.96(1) and kai = −0.81(1).
  • Forecasting MAE for φ⁴ model is 0.019 versus 0.023 for a simple rescaled‑mean baseline, indicating slightly better performance. predictive-accuracy
    Source sentence
    The mean absolute errors for the ϕ4 result and the rescaled mean prediction are 0.019 and 0.023, respectively. The ϕ4 result performs slightly better.
  • The scaling properties of weights and biases suggest exploitable market‑structure features for future investment strategies. theoretical
    Source sentence
    The scaling results for the weights and biases of the theory hint at the possibility of uncovering features of market structure, which once more fully explored may be able to be exploited.
  • The model could be extended to predict option prices or to design transaction‑cost‑minimizing trajectories. other
    Source sentence
    Finally, the model may have sufficient predictive power for more than one day ahead prediction and forecasts a full probability distribution, and so could be used to produce predictions for option prices, or in transaction cost minimizing trajectories [16].

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 42 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

https://github.com/dbachtis/phi4fin — the authors' own repository

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.