{"paper_id":"2504.17313","title":"Tokenizing Stock Prices for Enhanced Multi-Step Forecast and Prediction","authors":["Zhuohang Zhu","Haodong Chen","Qiang Qu","Xiaoming Chen","Vera Chung"],"paper_date":"2025-04-24","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2504.17313v1","page_url":"https://thequant.space/flowcharts/tokenizing-stock-prices-for-enhanced-multi-step-forecast-and/","scores":{"empirical_rigor":6.5,"math_complexity":7,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes the Patched Channel Integration Encoder (PCIE), a novel model for multi-step stock price forecasting and prediction that uses tokenization of price and price change channels.","quote":"we introduce a novel model, termed Patched Channel Integration Encoder (PCIE), to tackle both stock price forecasting and prediction.","summary":"Effective stock price forecasting (estimating future prices) and prediction (estimating future price changes) are pivotal for investors, regulatory agencies, and policymakers. These tasks enable informed decision-making, risk management, strategic planning, and superior portfolio returns. Despite…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Stock price data are non‑stationary with shifting means and variances over time.","type":"statistical","stated":"explicit","quote":"the non‑stationary distribution of stock prices, characterized by shifting means and variances over time, exacerbates the difficulty of accurate multi‑step forecasts."},{"assumption":"Forecasting targets follow a log‑normal distribution while prediction targets follow a normal distribution.","type":"statistical","stated":"explicit","quote":"Forecasting targets typically follow a log‑normal distribution, characterized by significant shifts in mean and variance over time, whereas prediction targets adhere to a normal distribution."},{"assumption":"Data are split into training, validation and testing sets with a 7:1:2 ratio.","type":"data","stated":"explicit","quote":"We split all data into training, validation and testing with a ratio of 7:1:2."},{"assumption":"Daily sampling interval is used for all price series.","type":"data","stated":"explicit","quote":"We collect the historical price from 2016/01/04 to 2023/12/29 with a daily sampling interval."},{"assumption":"Instance normalization can mitigate distribution shift in stock price data.","type":"model","stated":"explicit","quote":"... this technique coupled with our temporal learning block and patching techniques addresses the problem of distribution shift in stock price data."},{"assumption":"The benchmark models are tuned according to the recommendations of their original papers for a fair comparison.","type":"model","stated":"explicit","quote":"We tuned all baseline models according to the recommendations of their corresponding papers for a fair comparison."}]},"datasets":{"status":"available","items":[{"name":"US_71","provider":"Yahoo Finance","kind":"market-prices","access":"public","period":"2016/01/04 to 2023/12/29","universe":"71 high trade volume U.S. stocks across 9 major industries","quote":"For US_71, it contains the historical price of 71 high trade volume stocks from the U.S. stock market, representing the top 6-9 stocks in capital size and trade volume across the 9 major industries.","known_id":"yahoo-finance"},{"name":"US_14L","provider":"Yahoo Finance","kind":"market-prices","access":"public","period":"2005/01/04 to 2023/12/29","universe":"14 high market‑capitalization, high‑volume U.S. stocks","quote":"For US_14L, we selected 14 stocks characterized by high market capitalization and trading volume. We extended the time frame for data collection, capturing historical prices from 2005/01/04 to 2023/12/29, on a daily basis.","known_id":"yahoo-finance"}],"known":[{"id":"yahoo-finance","name":"Yahoo Finance","provider":"Yahoo","kind":"market-prices","access":"public","quote":"Both datasets are collected from Yahoo Finance and anomaly detection is performed to filter out the invalid data.","mentions":1,"used_in":1}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"daily"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R3","candidates":{"front_matter":"n/a","evidence_card":"n/a","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"forecast-accuracy","findings":[{"finding":"PCIE achieves the lowest MSE and MAE across all forecast horizons on both datasets, outperforming PatchTST, D‑Va, Autoformer, and Informer.","kind":"performance","quote":"The results obtained from our experiments are detailed in Table 2. The Patched Channel Integration"},{"finding":"The novel data preprocessing (combining price and price change) improves overall performance across all models by 0.6%–4%.","kind":"performance","quote":"the performance improvement detailed in Table 3 demonstrates that our novel data preprocessing method improves the performance across different models."},{"finding":"Ablation study shows that tokenization reduces MSE (e.g., from 0.0838 to 0.0690 for US_71 forecast horizon 10).","kind":"performance","quote":"Results summarized in Table 4 illustrate the critical role of tokenization in enhancing the model’s performance."},{"finding":"Models that ignore inter‑channel correlations (e.g., PatchTST) perform poorly on stock price forecasting.","kind":"negative","quote":"models like PatchTST, which convert the input into univariate series, fail to perform well because they omit these essential inter‑channel correlations."},{"finding":"Methods assuming stationary distributions (e.g., Autoformer) underperform on forecasting tasks where data distribution evolves.","kind":"negative","quote":"Autoformer employs a learned Fast Fourier Transform (FFT) attention mechanism, which assumes a stationary distribution. This assumption holds less effectively for forecasting tasks, where the data distribution evolves, resulting in subpar performance."}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":32,"citations":{"cited_by_in_corpus":1,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":1},"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":"2026-11-14T17:40:06Z"},"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/"}