{"paper_id":"2405.12993","title":"A novel portfolio construction strategy based on the core-periphery profile of stocks","authors":["Imran Ansari","Charu Sharma","Akshay Agrawal","Niteesh Sahni"],"paper_date":"2024-04-27","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2405.12993v1","page_url":"https://thequant.space/flowcharts/a-novel-portfolio-construction-strategy-based-on-the-core-pe/","scores":{"empirical_rigor":6,"math_complexity":7,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a portfolio construction strategy that selects stocks from the periphery of Planar Maximally Filtered Graphs based on core‑periphery structure.","quote":"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","summary":"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…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Pearson correlation coefficients adequately capture the relationships between stock returns for network construction.","type":"data","stated":"implied","quote":"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."},{"assumption":"The Planar Maximally Filtered Graph (PMFG) retains the most relevant links for the correlation‑based network.","type":"model","stated":"implied","quote":"Planar Maximally Filtered Graph: Planer maximally filtered graph (PMFG) is a network filtering approach introduced in41. It works by retaining the key representative links"},{"assumption":"Overlapping windows of 240 ticks (2 hours) with 75 % overlap provide a stable estimate of the network for high‑frequency data.","type":"data","stated":"implied","quote":"At each instant t, we construct the PMFG filtered network using 75% overlapping windows of size 240 ticks."},{"assumption":"Markowitz weights can be applied to the selected stocks, even though practical implementation may be costly.","type":"execution","stated":"explicit","quote":"theoretically applies to a large number of stocks, practical implementation can be challenging and costly"},{"assumption":"Uniform weighting of stocks is a feasible baseline for comparison.","type":"execution","stated":"explicit","quote":"Initially, portfolios are construct employing the uniform method, followed by a comprehensive comparison of the Sharpe ratios and average returns"},{"assumption":"Statistical significance of the core‑periphery profile is assessed via randomised networks preserving degree distribution.","type":"statistical","stated":"explicit","quote":"For each network we generate 100 randomized networks which preserve the same degree distribution as the original network"}]},"datasets":{"status":"available","items":[{"name":"NSE (India) 30‑second tick‑by‑tick data (CNX100)","provider":"National Stock Exchange of India","kind":"market-prices","access":"public","period":"January 2014 to December 2014","universe":"CNX100 index constituents","quote":"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."}],"known":[{"id":"the-graph","name":"The Graph (subgraphs)","provider":"The Graph","kind":"crypto-onchain","access":"public","quote":"Specifically, portfolios constructed from stocks within the periphery segment of the PMFG subgraph, exhibit superiority over portfolios constructed from the large hybrid measure (that is periphe","mentions":10,"used_in":3}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"unclear"},"transaction_costs":{"status":"available","modeled":"no","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R1","candidates":{"front_matter":"no","evidence_card":"no","evidence_card_quoted":false},"meaning":"a trading or execution result is reported without costs"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"Type‑1 (periphery‑based) portfolios consistently achieve higher Sharpe ratios and average returns than Types‑2‑5 and the market portfolio (Type‑6).","kind":"performance","quote":"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."},{"finding":"The core‑periphery profile is statistically significant in 77 % of the 686 two‑hour windows at the 5 % level.","kind":"statistical","quote":"77% of the 686 network windows were found to have a significant core‑periphery profile at 5% level of significance(𝑝‑value<0.05)"},{"finding":"Portfolios constructed from periphery stocks outperform traditional strategies and the market portfolio, delivering higher returns and lower risk.","kind":"performance","quote":"Our study reveals that our proposed core‑periphery‑based strategy consistently outperforms traditional"},{"finding":"Portfolios with Markowitz weights exhibit lower Sharpe ratios than those with uniform weights, highlighting practical challenges of the Markowitz method.","kind":"performance","quote":"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."},{"finding":"As portfolio size increases, the risk profile of Type‑1 converges toward that of the market portfolio (Type‑6).","kind":"performance","quote":"However, as the portfolio size increases, the risk profiles of Type‑1 become increasingly similar to that of Type‑6."}],"reported_performance":null,"out_of_sample":{"method":"walk-forward","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":39,"citations":{"cited_by_in_corpus":2,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":2},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":1,"extracted_at":"2026-11-14T18:10:32Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}