Tax Genius · India
Factor Research & Signal Validation
A research framework for testing investment factors and quantitative signals across definitions, universes, regimes, turnover, capacity and portfolio interaction.
Research framework published and architecture-reviewed:
Research framework
A factor should be defined economically and statistically before its historical premium is interpreted as a durable signal.
From idea to signal
Mission Saptanga separates the economic story from the mathematical implementation. The research record states why a variable might matter, how it is measured, when it becomes observable and how it enters a portfolio.
Robustness before ranking
A signal is challenged across reasonable definitions, time windows and universes before a preferred implementation is selected.
Portfolio interaction
Individual signal statistics are not enough. Correlation with other signals, concentration, turnover and regime dependence are examined at the portfolio layer.
Decision role
Validated signals can inform systematic research, but final use still depends on portfolio objectives, risk budgets, implementation constraints and governance state.
Methodology
- Specify the economic hypothesis and exact signal construction before performance review.
- Test alternative reasonable definitions rather than optimizing a single preferred specification.
- Evaluate turnover, crowding proxies, capacity and interaction with other signals.
- Assess behaviour across market regimes and portfolio combinations.
Evidence requirements
- Point-in-time or appropriately lagged data where required.
- Universe and rebalance rules.
- Transaction-cost and turnover assumptions.
- Cross-sectional and time-series robustness diagnostics.
Limitations
- Factor definitions vary across providers and academic literature.
- Observed historical premia can compress or reverse.
- No factor research result should be interpreted as a guaranteed excess-return source.