DATAENV / Alt Data /metadata /research_notes.md
Jitendra12421's picture
Upload Alt Data/metadata/research_notes.md
037a5b7 verified
|
raw
history blame
5.04 kB

Quant Research Notes

This note explains why the supplementary data in Alt Data was added and which papers or official data sources motivated it.

What was added

  • Alt Data/options/raw/
    • Original NIFTY and BANKNIFTY daily option-chain data moved out of the root folder.
  • Alt Data/options/processed/
    • Daily option features such as put-call ratios, open-interest changes, OI-weighted strikes, near-expiry ATM fields, and ATM straddle proxies.
  • Alt Data/external/raw/fred/
    • Daily external macro and cross-market series from FRED.
  • Alt Data/external/processed/external_daily_panel.csv
    • Aligned external panel with level, change, and return-style features where appropriate.
  • Alt Data/institutional/raw/
    • Daily institutional cash and F&O flow extracts plus NSE participant-wise derivatives archives.
  • Alt Data/institutional/processed/institutional_daily_panel.csv
    • Daily FII/DII cash buy-sell, index futures positioning, and index options positioning features.

Why these data types matter

  • Options open interest and option-chain structure:
    • Forward-looking positioning often shows up in put/call OI, OI changes, and near-expiry ATM behavior.
  • Institutional flow and participant positioning:
    • FII and DII cash flows capture who supplied or absorbed equity risk each day, while participant-wise index futures and index options positioning exposes directional buildup and hedge pressure.
  • Implied or volatility-linked information:
    • Volatility-sensitive features help both return and realized-volatility modeling.
  • Macro and financial conditions:
    • Rates, dollar strength, FX, and energy prices often improve regime detection and market-state modeling.
  • Global equity spillovers:
    • U.S. and Asia index moves commonly feed into next-session Indian index behavior.

Papers used to guide the additions

  1. Implied Volatility-Augmented GARCH Models: Evidence from the Indian Equity Market

  2. On options-driven realized volatility forecasting: Information gains via rough volatility model

  3. Nifty Index Options: Open Interest Analysis of Options Chain

  4. Forecasting Individual Stock Returns Using Macroeconomic and Technical Variables

  5. Stock prices and Macroeconomic indicators: Investigating a correlation in Indian context

Official data sources used

Practical use

  • Start with Data/processed/panels/daily_master_panel.csv for daily modeling.
  • Use Data/processed/features/<timeframe>/ for single-asset intraday experiments.
  • Join Alt Data/options/processed/*.csv into targeted option-flow studies when you want option-led predictors.
  • Join Alt Data/institutional/processed/institutional_daily_panel.csv when you want daily institutional flow and positioning context.