{"paper_id":"2512.02037","title":"Statistical Arbitrage in Polish Equities Market Using Deep Learning Techniques","authors":["Marek Adamczyk","Michał Dąbrowski"],"paper_date":"2025-11-20","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2512.02037v1","page_url":"https://thequant.space/flowcharts/statistical-arbitrage-in-polish-equities-market-using-deep-l/","scores":{"empirical_rigor":8.5,"math_complexity":7,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper introduces a deep‑learning based replication method using Long Short‑Term Memory networks for statistical arbitrage pairs trading in the Polish market.","quote":"as a new method that was not covered in the literature of the subject yet: considering Long short-term memory networks (LSTMs).","summary":"We study a systematic approach to a popular Statistical Arbitrage technique: Pairs Trading. Instead of relying on two highly correlated assets, we replace the second asset with a replication of the first using risk factor representations. These factors are obtained through Principal Components…","strategy":"PCA replication pairs trading"},"assumptions":{"status":"available","items":[{"assumption":"Stock price process is continuous and returns are normally distributed.","type":"model","stated":"explicit","quote":"For simplicity we are going to assume that such process is continuous. Then, for a time interval dt (f.e. 1/252 representing"},{"assumption":"Linear regression with constant beta decomposes returns into drift, market factor and idiosyncratic noise.","type":"model","stated":"explicit","quote":"which is a simple regression model (commonly called 2-parameters model) decomposing stock’s return into deterministic drift αdt, market (systematic) βFt factor and a stock-unique, mean 0, uncorrelated (idiosyncratic) process ϵt"},{"assumption":"Residuals follow an Ornstein‑Uhlenbeck mean‑reverting process.","type":"statistical","stated":"explicit","quote":"... Ornstein–Uhlenbeck process is going to be our natural candidate fulfilling all the requirements"},{"assumption":"Transaction cost of 0.1% per buy/sell is applied to all trades.","type":"execution","stated":"explicit","quote":"Transaction cost of c = 0.1% per buy/sell is going to be added."},{"assumption":"Backtesting uses 60 most influential companies from WIG20 and mWIG40.","type":"other","stated":"explicit","quote":"Applying techniques of Statistical Arbitrage presented by Avellaneda and Lee [4] to polish stock exchange required switching from around 500 stocks’ portfolios to just 60 most influential companies gathered by two main indices: WIG20 and mWIG40."}]},"datasets":{"status":"available","items":[{"name":"Polish equity constituents (WIG20 and mWIG40)","provider":"Warsaw Stock Exchange","kind":"market-prices","access":"public","period":"2017-2020 (backtesting) and 2019-2021 (L","universe":"60 most influential Polish companies","quote":"... to just 60 most influential companies gathered by two main indices: WIG20 and mWIG40."},{"name":"Sector indices (sWIG80 etc.)","provider":"Warsaw Stock Exchange","kind":"market-prices","access":"public","period":null,"universe":"sector‑specific Polish stocks","quote":"indices are way easier to replicate than WIG because of smaller components’ number."},{"name":"ETF based replication factors","provider":null,"kind":"alternative","access":"unknown","period":null,"universe":"Polish market ETFs","quote":"... considering Long short-term memory networks (LSTMs)."},{"name":"ING Bank Śląski stock data","provider":null,"kind":"market-prices","access":"unknown","period":"2019-2021 (training) and 2022 (testing)","universe":"single Polish bank stock","quote":"Let us now analyse how the stacked LSTM is working on real data- in the first example ING Bank Polski"}],"known":[],"sample":{"start":"2017","end":"2019","market":"Polish equities (WIG20, mWIG40)","asset_class":"equities","frequency":"daily"},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":null,"detail":"c = 0.1% per buy/sell (both opening and closing trades)","quote":"Transaction cost of c = 0.1% per buy/sell is going to be added.","source":"evidence_card","rule":"R2","candidates":{"front_matter":"unclear","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"c = 0.1% per buy/sell (both opening and closing trades)","card_quote":"Transaction cost of c = 0.1% per buy/sell is going to be added.","meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"PCA replication yields roughly 20% cumulative return and an annualized Sharpe ratio up to 2.63 over 2017‑2019.","kind":"performance","quote":"All methods yield profits in 2017-2019, with PCA achieving roughly 20 percent cumulative return and an annualized Sharpe ratio of up to 2.63."},{"finding":"During the 2020 COVID‑19 recession only the ETF‑based approach remains profitable (~5% annual return), while PCA and LSTM underperform.","kind":"performance","quote":"During the COVID-19 recession, only the ETF based approach remains profitable (about 5 percent annual return), while PCA and LSTM methods underperform."},{"finding":"LSTM overall portfolio Sharpe ratios are 0.60 (2017), 2.09 (2018) and –1.53 (2019), with a final return around 10%, worse than PCA.","kind":"performance","quote":"... The final return is around 10% which is worse than with both PCA approaches."},{"finding":"LSTM sector‑specific Sharpe ratios vary, with highest values in 2018 and many negative ratios in 2019.","kind":"performance","quote":"LSTM is underperforming in 2019 for most of the groups. Highest ratios were achieved in 2018 with 2017 giving more than half of positive ratios."}],"reported_performance":{"strategy":"PCA replication pairs trading","sharpe":2.63,"sharpe_benchmark":null,"annual_return_pct":null,"total_return_pct":20,"max_drawdown_pct":null},"out_of_sample":{"method":"holdout","headline_out_of_sample":true}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":41,"citations":{"cited_by_in_corpus":0,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":0},"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":"2026-11-14T17:21:32Z"},"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/"}