Home › Research and editorial standards › Data Directory

RPS Data Directory and Citation Guide

This is the normative entry point for the current market-level aggregate data on this site. Visible tables, machine-readable JSON, and Dataset metadata all come from the same completed snapshot, avoiding conflicts in dates between page displays and structured data.

Current Version: 2026-09-24 · Valid RPS50 Samples: 20,632 · Free Public Access

5,212A-Shares · 49 Industries
8,948U.S. Stocks · 48 Industries
2,774Hong Kong Stocks · 169 Industries
3,698Japanese Stocks · 33 Industries

Latest Aggregate Snapshot

MarketValid securitiesIndustry classificationCompletion Update Time
China A-shares5,212492026-09-24 11:08:28
US stocks8,948482026-09-25 13:40:38
Hong Kong stocks2,7741692026-09-25 12:58:24
Japanese stocks3,698332026-09-25 13:11:17
Total20,632Sample pool for four independently ranked markets

Open Machine-Readable JSON

What the Dataset Contains

Public distribution includes the number of valid securities for which RPS50 can be calculated in each market at the current time, the number of industry classifications, and the data completion time, as well as Dataset metadata describing the aggregate statistics. It does not include tick-by-tick market data, account data, personalized advice, or future earnings predictions.

“Valid securities” are used in the denominator calculation and do not equal the total number of securities listed on the exchange. The corresponding window requires sufficient historical length and valid positive closing prices; new listings, suspensions, delistings, missing data, and quality checks may all affect the quantity.

Calculation and Sources

Aggregated facts come from read-only database queries after data updates are completed on the site. Each market is calculated independently: first, obtain the closing price range returns for the specified trading day window, then perform percentile ranking within the market on non-empty returns. Tied returns use the highest rank within the tied group, and scores are kept to two decimal places. Detailed implementation see RPS calculation methodology。

Version and Reproducibility

JSON filenames remain stable to allow research systems to continuously read the latest completed snapshots; among them `dateModified`, each market's `updatedAt`, and the tables visible on this page are updated synchronously. Historical weekly reports retain their respective observation dates and will not be overwritten by the current snapshot. When historical context is needed, the original weekly report or Cross-Week Statistical Trends。

Suggested Citation Format

RPS Stock (2026-09-24 completed snapshot), four markets RPS50 valid samples 20,632 only: A-shares 5,212 US stocks 8,948 Hong Kong stocks 2,774 Japanese stocks 3,698. Source: https://rps001.alaala.xyz/research/rps-data-catalog.html. RPS is the historical performance percentile within each market.

Usage and Interpretation Boundaries

What Can Be Done

Verify the current ranking denominator, compare sample sizes across markets, record research snapshots, and provide sources for method or report citations.

What Cannot Be Inferred

Sample size cannot be interpreted as market rise or fall, the same RPS cannot be treated as the same absolute return across markets, and high scores cannot be rewritten as the probability of rise.

Related Authoritative Pages

RPS calculation methodology: Exact return formula and percentile rulesFour-Market Data Coverage Archive: Securities pool, industry, and update boundariesResearch and editorial standards: Evidence, revisions, and citation standardsEnglish data standards: English methods, error correction, and citation boundariesEnglish score notes: Daily multi-period evidence recordsToday's RPS Market Pulse: Current strong sample distributionEarliest Complete Weekly Report: Retain historical cross-section 2026-07-20Weekly Report Archiving: Historical cross-section with observation date

Page generation: 2026-09-25; public facts snapshot: 2026-09-24.