JaySense Research Methodology

What is the JaySense Research Score?

The JaySense Research Score is a quantitative research metric designed to help users evaluate multiple characteristics of an Indian listed security within a consistent, repeatable framework.

The score is computed daily — after market close — for each stock in the research universe. It reflects conditions as of that day's available data and is not a prediction of future price movements.

Factors evaluated

The engine evaluates a stock across multiple dimensions. The table below describes what is evaluated — not the specific indicators, weights, or thresholds used. Individual factor contributions vary depending on data availability and market conditions.

The relative contribution of individual factors is determined by the research engine and is not published. JaySense periodically evaluates and adjusts factor construction using historical data. Factors with insufficient or stale data may receive a neutral weighting for that stock on that day.

How to interpret the score

Scores are grouped into five research bands to assist interpretation:

Scores are computed from end-of-day data. An intraday staleness indicator is shown when price has moved materially from the prior close, to signal that the score may not reflect current conditions.

Historical validation

JaySense conducts periodic historical validation of its research scores. The validation process samples historical (stock, date) pairs, applies the research engine to conditions as of that date, and measures the forward price outcome over multiple holding periods (7, 15, and 20 trading days).

The key question asked is: do higher-scored stocks tend to have better average forward outcomes than lower-scored stocks, across the full sample? If so, the engine shows statistical edge. If not, the methodology is reviewed.

Validation is not a guarantee. Historical patterns may not persist. A validation result showing positive historical edge does not imply future performance will replicate that edge.

Data sources

JaySense uses the following categories of public data:

• Price and volume data — end-of-day OHLCV from financial data providers
• Delivery volume — publicly available NSE equity delivery data
• Fundamental data — publicly available financial statements, ratios, and valuation metrics
• Shareholding patterns — quarterly NSE/BSE shareholding disclosures (promoter, FII, DII, public)
• Quarterly financials — company-reported quarterly results from public filings
• FII/DII flows — daily institutional net flow data published by NSE
• Corporate announcements — NSE corporate announcements (results, dividends, AGM, etc.)
• Market news — publicly available market news from RSS and public sources
• Options OI data — publicly available NSE options open interest data

JaySense does not use non-public information, insider data, or proprietary data feeds not available to the general public.

Data refresh & availability

Research scores and underlying data are refreshed on the following approximate schedule on trading days:

Data availability varies by stock. Scores for stocks with limited data history or stale fundamental data carry a data quality indicator. Scores computed from fewer than 30 days of price history may not be representative.

Limitations

Users should understand the following limitations before acting on any research output from JaySense:

• Historical basis. All scores are computed from historical and currently available data. Historical relationships between scores and outcomes may not persist in future market conditions.
• EOD data only. Scores reflect end-of-day conditions. Intraday price movements, news events, or corporate announcements that occur after a score is computed are not reflected until the next scheduled update.
• Data gaps. Fundamental data for some stocks may be incomplete, stale, or unavailable. Scores for such stocks carry a data quality indicator and should be interpreted accordingly.
• No macro awareness. The research engine does not model macroeconomic factors, geopolitical events, regulatory changes, or monetary policy shifts. These can materially affect market outcomes independent of stock-level scores.
• Liquidity not modelled. Scores are computed without regard to a user's position size. A high-conviction score for an illiquid stock does not indicate that the position can be entered or exited at the modelled price.
• No insider or non-public information. JaySense uses only publicly available data. Events known to insiders but not yet publicly disclosed are not captured.
• Factor stability. Research engine factors and their construction are reviewed periodically. Past validation results may not apply to the current version of the engine.
• Small-cap and illiquid stocks. The engine is calibrated primarily on liquid mid- and large-cap NSE-listed securities. Research scores for illiquid or thinly traded stocks may be less reliable.

What JaySense does NOT do

Risk disclosure

For full risk disclosure, see Risk Disclosure · Disclaimer · Terms of Service