Paper: arXiv 2405.12993
Authors: Imran Ansari, Charu Sharma, Akshay Agrawal, Niteesh Sahni
Abstract
This paper highlights the significance of mesoscale structures, particularly the core-periphery structure, in financial networks for portfolio optimization. We build portfolios of stocks belonging to the periphery part of the Planar maximally filtered subgraphs of the underlying network of stocks created from Pearson correlations between pairs of stocks and compare its performance with some well-known strategies of Pozzi et. al. hinging around the local indices of centrality in terms of the Sharpe ratio, returns and standard deviation. Our findings reveal that these portfolios consistently outperform traditional strategies and further the core-periphery profile obtained is statistically significant across time periods. These empirical findings substantiate the efficacy of using the core-periphery profile of the stock market network for both inter-day and intraday trading and provide valuable insights for investors seeking better returns.
Complexity vs Empirical Score
- Math Complexity: 7.0/10
- Empirical Rigor: 6.0/10
- Quadrant: Holy Grail — high math complexity, high empirical rigor
Why this score: The paper employs advanced mathematical concepts from graph theory and network science (e.g., PMFG, Rombach and Rossa algorithms for core-periphery detection, Markowitz optimization) for portfolio construction. It demonstrates strong empirical rigor by backtesting the strategy on both daily and high-frequency (30-second) data for an Indian stock market, using standard performance metrics (Sharpe ratio, returns, standard deviation) and statistical significance tests, which makes it highly data and implementation-heavy.
Research Flowchart
flowchart TD
A["Research Goal: Test Portfolio<br>Performance based on<br>Core-Periphery Structure"] --> B["Data Inputs:<br>Stock Prices (Inter-day & Intraday)"]
B --> C["Network Construction:<br>Planar Maximally Filtered Graph<br>(Pearson Correlations)"]
C --> D["Topology Analysis:<br>Identify Core vs. Periphery<br>Nodes"]
D --> E["Portfolio Strategy:<br>Construct Portfolio using<br>Peripheral Stocks"]
E --> F["Performance Benchmarking:<br>Sharpe Ratio, Returns, Std. Deviation<br>vs. Traditional Strategies"]
F --> G{"Key Outcomes"}
G --> H["Periphery Portfolios<br>Outperform Traditional Ones"]
G --> I["Core-Periphery Profile<br>is Statistically Significant"]
G --> J["Effective for<br>Inter-day & Intraday Trading"]
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 portfolio construction strategy that selects stocks from the periphery of Planar Maximally Filtered Graphs based on core‑periphery structure.
Source sentence
We build portfolios of stocks belonging to the periphery part of the Planar maximally filtered subgraphs of the underlying network of stocks created from Pearson correlations between pairs of stocks
Assumes 6
- Pearson correlation coefficients adequately capture the relationships between stock returns for network construction. data
Source sentence
We compute 𝑎𝑖𝑗 as 1+𝜌𝑖𝑗 2 when 𝑖 ≠ 𝑗 and 𝑎𝑖𝑗= 0 otherwise. Here 𝜌𝑖𝑗 is the Pearson correlation coefficient between vertices 𝑖 and 𝑗.
- The Planar Maximally Filtered Graph (PMFG) retains the most relevant links for the correlation‑based network. model
Source sentence
Planar Maximally Filtered Graph: Planer maximally filtered graph (PMFG) is a network filtering approach introduced in41. It works by retaining the key representative links
- Overlapping windows of 240 ticks (2 hours) with 75 % overlap provide a stable estimate of the network for high‑frequency data. data
Source sentence
At each instant t, we construct the PMFG filtered network using 75% overlapping windows of size 240 ticks.
- Markowitz weights can be applied to the selected stocks, even though practical implementation may be costly. execution
Source sentence
theoretically applies to a large number of stocks, practical implementation can be challenging and costly
Show 2 more
- Uniform weighting of stocks is a feasible baseline for comparison. execution
Source sentence
Initially, portfolios are construct employing the uniform method, followed by a comprehensive comparison of the Sharpe ratios and average returns
- Statistical significance of the core‑periphery profile is assessed via randomised networks preserving degree distribution. statistical
Source sentence
For each network we generate 100 randomized networks which preserve the same degree distribution as the original network
Data
- NSE (India) 30‑second tick‑by‑tick data (CNX100) market-prices public
Source sentence
We conduct our analysis using 30-second tick-by-tick high-frequency data obtained from the NSE (India), covering the period from January 2014 to December 2014, comprising a total of 164,640 ticks.
The Graph (subgraphs)
Costs
Not modeled
Finds 5
- Type‑1 (periphery‑based) portfolios consistently achieve higher Sharpe ratios and average returns than Types‑2‑5 and the market portfolio (Type‑6). performance
Source sentence
Supplementary Figs. S8‑S9 illustrate that Type‑1 exhibits a significant outperformance compared to Types‑2, 3, 4, 5, and 6 in terms of both the Sharpe ratio and average returns.
- The core‑periphery profile is statistically significant in 77 % of the 686 two‑hour windows at the 5 % level. statistical
Source sentence
77% of the 686 network windows were found to have a significant core‑periphery profile at 5% level of significance(𝑝‑value<0.05)
- Portfolios constructed from periphery stocks outperform traditional strategies and the market portfolio, delivering higher returns and lower risk. performance
Source sentence
Our study reveals that our proposed core‑periphery‑based strategy consistently outperforms traditional
- Portfolios with Markowitz weights exhibit lower Sharpe ratios than those with uniform weights, highlighting practical challenges of the Markowitz method. performance
Source sentence
Finally, our study reveals that portfolios with Markowitz weights exhibit a lower Sharpe ratio compared to those with uniform weights, aligning with the classical framework of the Markowitz method.
- As portfolio size increases, the risk profile of Type‑1 converges toward that of the market portfolio (Type‑6). performance
Source sentence
However, as the portfolio size increases, the risk profiles of Type‑1 become increasingly similar to that of Type‑6.
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.
2 papers in the archive
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 (45)
- Detection of core–periphery structure in networks using spectral methods and geodesic paths (2016) — link
- Density-based and transport-based core-periphery structures in networks (2014) — link
- A Simple Model of Capital Market Equilibrium with Incomplete Information (1987) — link
- Spread of risk across financial markets: better to invest in the peripheries (2013) — link
- Cluster analysis for portfolio optimization (2007) — link
- Self-Affine Fractals and Fractal Dimension (1985) — link
- Collective transport of coupled particles (2012) — link
- Overlapping Communities Explain Core–Periphery Organization of Networks (2014) — link
and 37 more
- Profiling core-periphery network structure by random walkers (2013) — link
- Models of core/periphery structures (2000) — link
- A survey on nature inspired metaheuristic algorithms for partitional clustering (2014) — link
- Dynamical correlations in financial systems (2007) — link
- A tool for filtering information in complex systems (2005) — link
- Clustering Indian stock market data for portfolio management (2010) — link
- The formation of a core-periphery structure in heterogeneous financial networks (2020) — link
- Determinants of Portfolio Performance II: An Update (1991) — link
- Evolutionary Optimization of Constrained $k$-Means Clustered Assets for Diversification in Small Portfolios (2009) — link
- Optimization Methods in Finance (2006) — link
- Hierarchical structure in financial markets (1999) — link
- Global Portfolio Optimization (1992) — link
- Performance Measurement in a Downside Risk Framework (1994) — link
- Long-Term Storage Capacity of Reservoirs (1951) — link
- Dynamic Portfolio Strategy Using Clustering Approach (2017) — link
- Finding multiple core-periphery pairs in networks (2017) — link
- Core-Periphery Structure in Networks (Revisited) (2017) — link
- Nonlinear dependencies on Brazilian equity network from mutual information minimum spanning trees (2019) — link
- Stock market as temporal network (2018) — link
- The Fractal Geometry of Nature. (1984) — link
- Development of stock correlation networks using mutual information and financial big data (2018) — link
- Portfolio optimization based on network topology (2018) — link
- Network analysis of the Chinese stock market during the turbulence of 2015–2016 using log-returns, volumes and mutual information (2019) — link
- Portfolio optimization based on empirical mode decomposition (2019) — link
- Introduction to Risk Parity and Budgeting (2013) — link
- Mutual information based stock networks and portfolio selection for intraday traders using high frequency data: An Indian market case study (2019) — link
- Illiquidity and stock returns: cross-section and time-series effects (2002) — link
- All That Glitters: The Effect of Attention and News on the Buying Behavior of Individual and Institutional Investors (2007) — link
- Limited attention, information disclosure, and financial reporting (2003) — link
- A Network Approach to Portfolio Selection (2014) — link
- Interbank tiering and money center banks (2014) — link
- A mutual information based R-vine copula strategy to estimate VaR in high frequency stock market data (2021) — link
- Efficient Capital Markets: A Review of Theory and Empirical Work (1970) — link
- PORTFOLIO SELECTION* (1952) — link
- Inefficient Markets (2000) — link
- Matrix Analysis and Applied Linear Algebra, Second Edition (2023) — link
- Exploring the core–periphery and community structure in the financial networks through random matrix theory (2025) — link
0 of 45 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.