Paper: arXiv 2601.03974
Authors: Anubha Goel, Amita Sharma, Juho Kanniainen
Abstract
Topological Data Analysis (TDA), an emerging field in investment sciences, harnesses mathematical methods to extract data features based on shape, offering a promising alternative to classical portfolio selection methodologies. We utilize persistence landscapes, a type of summary statistics for persistent homology, to capture the topological variation of returns, blossoming a novel concept of ``Topological Risk". Our proposed topological risk then quantifies portfolio risk by tracking time-varying topological properties of assets through the \(L_p\) norm of the persistence landscape. Through optimization, we derive an optimal portfolio that minimizes this topological risk. Numerical experiments conducted using nearly a decade long S&P 500 data demonstrate the superior performance of our TDA-based portfolios in comparison to the seven popular portfolio optimization models and two benchmark portfolio strategies, the naive \(1/N\) portfolio and the S&P 500 market index, in terms of excess mean return, and several financial ratios. The outcome remains consistent through out the computational analysis conducted for the varying size of holding and investment time horizon. These results underscore the potential of our TDA-based topological risk metric in providing a more comprehensive understanding of portfolio dynamics than traditional statistical measures. As such, it holds significant relevance for modern portfolio management practices.
Complexity vs Empirical Score
- Math Complexity: 8.5/10
- Empirical Rigor: 6.0/10 (capped from 7.0: transaction costs not modelled; see penalty flags)
- Quadrant: Holy Grail — high math complexity, high empirical rigor (effective empirical rigor capped from 7.0)
Why this score: The paper introduces advanced Topological Data Analysis (TDA) concepts like persistent homology and persistence landscapes, requiring significant mathematical abstraction, while the empirical section details nearly a decade of S&P 500 backtesting against multiple baselines with standard financial metrics.
Research Flowchart
flowchart TD A["Research Goal<br>Are topological portfolios superior<br>to classical portfolios?"] --> B["Data Input<br>S&P 500 Historical Data"] B --> C["Methodology: Topological Data Analysis<br>Calculate Persistence Landscapes<br>Quantify Topological Risk via Lp norm"] C --> D["Computation: Optimization<br>Minimize Topological Risk<br>Construct Optimal Portfolio"] D --> E["Outcome: Validation<br>Compare against 7 classical models<br>and 2 benchmarks"] E --> F["Findings<br>TDA portfolios yield higher excess returns<br>and superior financial ratios<br>across varying horizons"]
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 Topological Risk metric based on persistence landscapes and an optimization model (TDA-PO) that minimizes this risk to construct portfolios.
Source sentence
We utilize persistence landscapes, a type of summary statistics for persistent homology, to capture the topological variation of returns, blossoming a novel concept of ``Topological Risk".
Assumes 2
- Zero cross‑interaction among distinct assets in the topological risk matrix. model
Source sentence
we operate under the assumption of zero cross-interaction among distinct assets
- Daily returns are computed from closing prices without accounting for slippage or market impact. execution
Source sentence
The daily return for i-th stock at t-th day is calculated as rit = pc_it−pc_it−1 pc_it−1
Data
- S&P 500 daily closing prices and constituents market-prices commercial
Source sentence
The sample data contains daily closing prices of the S&P500 and its constituents over a sample period spanning nearly 10 years from December 10, 2012, to August 11, 2022
Costs
Unclear - sources disagree
Source sentence (the authors' words)
The returns have been adjusted to account for a transaction cost rate of 0.3%.
Finds 3
- TDA‑PO achieves the highest excess mean return (EMR), almost double the mean return of variance‑based models. performance
Source sentence
The proposed model, TDA‑PO attains the highest value of mean return in comparison to all the other seven models and both the benchmark portfolio strategies. The model TDA‑PO attains almost double mean return in comparison to all the three variance‑based models
- TDA‑PO consistently ranks first across all risk‑adjusted return metrics (Sharpe, Sortino, Rachev, etc.). performance
Source sentence
One of the most striking results from Table 4 is that TDA‑PO consistently ranks first in all risk‑adjusted return metrics when the second rank achieved by the TDA‑IPO.
- TDA‑PO and its constrained variants exhibit lower turnover than most benchmark models, indicating more stable allocations. performance
Source sentence
...) exhibit relatively lower turnover compared to benchmarked PO models such as Sharpe, STARR, MVaR and Omega
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 29 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.
Cited References (171)
- Topological data analysis in investment decisions (2020)
- Universality of persistence diagrams and the bottleneck and Wasserstein distances (2022)
- Optimization of conditional value-at risk (2000) — link
- Liquidity and stock returns (1986)
- The measurement and control of trading costs (1990)
- Dynamic rebalancing portfolio models with analyses of investor sentiment (2022)
- Using value-at-risk to control risk taking: how wrong can you be? (1998)
- Robust portfolio selection problems (2003)
and 163 more
- A robust mean absolute deviation model for portfolio optimization (2011)
- Can topological transitions in cryptocurrency systems serve as early warning signals for extreme fluctuations in traditional markets? (2025) — link
- Why topological data analysis detects financial bubbles? — → 2304.06877
- Uncertainty, volatility and the persistence norms of financial time series (2023) — link
- Topological data analysis of critical transitions in financial networks (2017)
- Topological features of multivariate distributions: Dependency on the covariance matrix (2021)
- Topological recognition of critical transitions in time series of cryptocurrencies (2020)
- Trading is hazardous to your wealth: The common stock investment performance of individual investors (2000)
- Risk Reduced Sparse Index Tracking Portfolio: A Topological Data Analysis Approach (2023)
- Active vs. Passive management: New evidence from exchange traded funds (2009)
- Clustering of financial time series with application to index and enhanced index tracking portfolio (2005)
- High-dimensional index tracking with cointegrated assets using an hybrid genetic algorithm (2011)
- Variable selection with big data based on zero norm and via sequential monte carlo (2019)
- Optimal hedging using cointegration (1999)
- Sparse portfolio selection via topological data analysis based clustering — → 2401.16920
- Using genetic algorithm to support portfolio optimization for index fund management (2005)
- Sparse and stable M arkowitz portfolios (2009)
- Regular (ized) hedge fund clones (2010)
- Tracking hedge funds returns using sparse clones (2018)
- A generalized approach to portfolio optimization: Improving performance by constraining portfolio norms (2009)
- Optimal cardinality constrained portfolio selection (2013)
- Vast portfolio selection with gross-exposure constraints (2012)
- Regression shrinkage and selection via the lasso (1996)
- A hybrid optimization approach to index tracking (2009)
- Minimizing the tracking error of cardinality constrained portfolios (2018)
- An efficient optimization approach for a cardinality-constrained index tracking problem (2016)
- Optimal portfolio selection using regularization (2011)
- Solving norm constrained portfolio optimization via coordinate-wise descent algorithms (2014)
- Constructing optimal sparse portfolios using regularization methods (2015)
- Sparse weighted-norm minimum variance portfolios (2016)
- Robust minimum variance portfolio with $L$-infinity constraints (2014)
- Sparse index clones via the sorted $\ell_1$-Norm (2022)
- Sparse index tracking using sequential Monte Carlo (2022)
- Barcodes: the persistent topology of data (2008)
- Simultaneous pursuit of out-of-sample performance and sparsity in index tracking portfolios (2013)
- Sparse portfolios for high-dimensional financial index tracking (2017)
- Sparse portfolio selection via the sorted $\ell_1$-Norm (2020)
- Simultaneous regression shrinkage, variable selection, and supervised clustering of predictors with OSCAR (2008)
- Subsampling methods for persistent homology (2015)
- Topological Data Analysis for Identifying Critical Transitions in Cryptocurrency Time Series (2020)
- Stochastic convergence of persistence landscapes and silhouettes (2014)
- An Introduction to Topological Data Analysis: fundamental and practical aspects for data scientists (2017) — link
- The persistence landscape and some of its properties (2018) — link
- Empirical study of financial crises based on topological data analysis (2020)
- Practical portfolio performance measurement and attribution (2023)
- The adaptive L asso and its oracle properties (2006)
- On the adaptive elastic-net with a diverging number of parameters (2009)
- Robust topological inference: Distance to a measure and kernel distance (2017)
- Nonnegative-lasso and application in index tracking (2014)
- Lasso-based index tracking and statistical arbitrage long-short strategies (2020)
- High-dimensional index tracking based on the adaptive elastic net (2020)
- Nonnegative elastic net and application in index tracking (2014)
- Nonnegative adaptive lasso for ultra-high dimensional regression models and a two-stage method applied in financial modeling (2016)
- Weighted elastic net penalized mean-variance portfolio design and computation (2015)
- Detecting strange attractors in uid turbulence (1981)
- Statistical topological data analysis using persistence landscapes. (2015)
- Chatter detection in turning using persistent homology (2016)
- A roadmap for the computation of persistent homology (2017)
- Analyzing collective motion with machine learning and topology (2019)
- A bayesian framework for persistent homology (2020)
- SW1PerS: Sliding windows and 1-persistence scoring; discovering periodicity in gene expression time series data (2015)
- Topological data analysis of contagion maps for examining spreading processes on networks (2015)
- Harnessing the power of topological data analysis to detect change points (2020)
- Modeling the spread of the Zika virus using topological data analysis (2018)
- Refining understanding of corporate failure through a topological data analysis mapping of Altman’s Z-score model (2020)
- Why topological data analysis detects financial bubbles? — → 2304.06877
- Topology and data (2009)
- Using persistent homology as preprocessing of early warning signals for critical transition in flood (2021)
- Topological characterization and early detection of bifurcations and chaos in complex systems using persistent homology (2017)
- Investigation of flash crash via topological data analysis (2021)
- Topological methods for data modelling (2020)
- Topological data analysis (2018)
- Sliding windows and persistence: An application of topological methods to signal analysis (2015)
- Persistent topological features of dynamical systems (2016)
- Block sparse variational Bayes regression using matrix variate distributions with application to SSVEP detection (2020)
- Greedy projected gradient-Newton method for sparse logistic regression (2019)
- Supervised feature selection with orthogonal regression and feature weighting (2020)
- Semisupervised feature selection via spline regression for video semantic recognition (2014)
- An introduction to variable and feature selection (2003)
- Short-term sparse portfolio optimization based on alternating direction method of multipliers (2018)
- Sparse Index Tracking With K-Sparsity or $\varepsilon$-Deviation Constraint via $\ell_0$-Norm Minimization (2022)
- Low-rank plus sparse decomposition of covariance matrices using neural network parametrization (2021)
- The better turbulence index? Forecasting adverse financial markets regimes with persistent homology (2021)
- Extracting insights from the shape of complex data using topology (2013)
- Generating dynamical neuroimaging spatiotemporal representations (DyNeuSR) using topological data analysis (2019)
- Perslay: A neural network layer for persistence diagrams and new graph topological signatures (2020)
- Topological machine learning for multivariate time series (2022)
- Why topology for machine learning and knowledge extraction? (2018)
- Learning topology: bridging computational topology and machine learning (2021)
- A survey of topological machine learning methods (2021)
- Persistent homology for time series and spatial data clustering (2015)
- Index tracking and enhanced indexing using mixed conditional value-at-risk (2018)
- Topological data analysis in biomedicine: A review (2022) — link
- Segmentation of biomedical images based on a computational topology framework (2020) — link
- Topological Data Analysis of Single-Trial Electroencephalographic Signals (2018)
- Reviews: Topological Distances and Losses for Brain Networks (2021) — link
- Topological Data Analysis (TDA) for Time Series (2019) — link
- Extracting insights from the shape of complex data using topology (2013)
- Identification of type 2 diabetes subgroups through topological analysis of patient similarity (2015)
- Topological data analysis for discovery in preclinical spinal cord injury and traumatic brain injury (2015)
- Using multidimensional topological data analysis to identify traits of hip osteoarthritis (2018)
- Lung topology characteristics in patients with chronic obstructive pulmonary disease (2018)
- Modeling the spread of the zika virus using topological data analysis (2018)
- Protein-folding analysis using features obtained by persistent homology (2020)
- Sliding windows and persistence: An application of topological methods to signal analysis (2015)
- Alpha, Betti and the Megaparsec Universe: on the Topology of the Cosmic Web (2011)
- Persistence Barcodes for Shapes (2005) — link
- Persistence-Based Structural Recognition (2014)
- Homological sensor networks (2007)
- Evasion paths in mobile sensor networks (2015)
- Persistent-Homology-based Machine Learning and its Applications – A Survey (2018)
- Automatic recognition and tagging of topologically different regimes in dynamical systems (2014)
- Non-random topology of stock markets (2001)
- Portfolio Concentration and the Geometry of Co-Movement (2013)
- Clustering and classification of time series using topological data analysis with applications to finance (2020)
- A Short Survey of Topological Data Analysis in Time Series and Systems Analysis (2018)
- Can TDA Be a New Risk Measure? An Application to Finance of Persistent Homology (2016)
- Deterministicky chaos a jeho fyzikalni aplikace (2003)
- Utilizing topological data analysis for studying signals of time-delay systems (2017)
- Time Series Featurization via Topological Data Analysis (2019) — link
- Persistence images: A stable vector representation of persistent homology (2017)
- Portfolio selection (1952) — link
- Risk reduction in large portfolios: Why imposing wrong constraints helps (2003) — link
- Improved estimation of the covariance matrix of stock returns with an application to portfolio selection (2003) — link
- A well-conditioned estimator for large-dimensional covariance matrices (2004) — link
- A generalized approach to portfolio optimization: Improving performance by constraining portfolio norms (2009) — link
- Sparsity and stability for minimum-variance portfolios (2022) — link
- Minimum Variance Portfolio Composition (2010)
- Putting Markowitz theory to work (1981)
- Bayesian and CAPM estimators of the means: Implications for portfolio selection (1991)
- The effect of errors in means, variances, and covariances on optimal portfolio choice (1993)
- Perturbation analysis of optimization problems (2000)
- Cardinality-constrained portfolio optimization with short selling and risk-neutral interest rate (2021) — link
- Sparse and robust portfolio selection via semi-definite relaxation (2020) — link
- A mean-absolute deviation-skewness portfolio optimization model (1993)
- Generalized quantiles as risk measures (2014)
- The sharpe ratio (1994)
- Optimizing the omega ratio using linear programming (2014)
- Optimal versus naive diversification: How inefficient is the 1/N portfolio strategy? (2009)
- Optimizing Omega (2006) — link
- 60 years of portfolio optimization: Practical challenges and current trends (2014)
- Can robust portfolio optimisation help to build better portfolios? (2007)
- A numerically stable dual method for solving strictly convex quadratic programs (1983)
- Cardinality versus q-norm constraints for index tracking (2014)
- Stochastic dominance, mean variance, and Gini’s mean difference (1982)
- On consistency of stochastic dominance and mean- semideviations models (2001)
- Risk Measurement: An Introduction to Value at Risk (1996)
- Conditional value-at-risk for general loss distributions (2002)
- 60 Years of portfolio optimization: Practical challenges and current trends (2014)
- Portfolio selection models: a review and new directions (2009)
- Persistent homology of tumor CT scans is associated with survival in lung cancer (2021)
- Applied nonlinear time series analysis: applications in physics, physiology and finance (2005)
- Topological data analysis of financial time series: Landscapes of crashes (2018)
- Putting markowitz theory to work (1981)
- Efficient asset management: A practical guide to stock portfolio optimization and asset allocation (2023)
- Asset equilibrium: Combining investor views with market equilibrium (1991)
- Asset Allocation and Risk Management (2009)
- On the Sensitivity of Mean-Variance-Efficient Portfolios to Changes in Asset Means: Some Analytical and Computational Results (2015)
- Testing for independence of large dimensional vectors (2019)
- Detecting Stress from Multivariate Time Series Data Using Topological Data Analysis (2023)
- On time-series topological data analysis: New data and opportunities (2016)
- A persistent-homology-based turbulence index & some applications of TDA on financial markets (2022) — link
- Equivalence of Robust VaR and CVaR Optimization (2016) — link
3 of 171 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.