Too Big to Co-Move
Concentrated sectors co-move less with the rest of the market: a one-point rise in top-three share lowers monthly R² by 0.75 points (permutation p = 0.04), though concentration does not explain why markets disconnected.
Research / VolEdge: The Variance Risk Premium
VolEdge: Measuring and Trading the Variance Risk Premium with Model-Free Risk-Neutral Moments
The gap between what options price and what the underlying realizes is the variance risk premium. Can it be measured model-free, and does it pay to trade?
Key results
0.85 vs 0.68
Sharpe of the GARCH volatility-targeted SPY/BIL sleeve vs SPY, net of costs (evaluation 2020 to 2024, including Covid)
-19.7% vs -33.7%
maximum drawdown, a 41% reduction
-0.16
Sharpe of naive short-variance harvesting: the premium pays for crash risk, it is not free yield
0.000 pp
difference between the backtest engine and buy-and-hold at any rebalance frequency
Abstract
VolEdge is a full-stack research platform built around a single thesis: the gap between what options price and what the underlying realizes is the risk premium, and that gap can be measured and traded. The platform extracts the risk-neutral distribution model-free via Bakshi-Kapadia-Madan (BKM) moments and compares it, horizon-matched, against the physical distribution estimated from realized returns; the spread Q minus P is surfaced for variance, skew, and kurtosis. A no-look-ahead walk-forward engine then backtests premium-driven strategies. We report two honest results on free data (2019 to 2024, net of modeled costs): a GARCH(1,1) volatility-targeted strategy beat SPY buy-and-hold on a risk-adjusted basis (Sharpe 0.85 vs. 0.68) with 41% lower maximum drawdown, while a naive long-only variance-premium harvest lost money (Sharpe -0.16), empirically confirming the premium as compensation for crash risk rather than harvestable alpha. The BKM extractor is validated against Black-Scholes chains of known volatility, and the backtest engine reproduces buy-and-hold to 0.000 pp at any rebalance frequency.
Methods
- Bakshi-Kapadia-Madan model-free risk-neutral variance, skew and kurtosis from live option chains
- Implied volatility and Greeks computed locally (Black-Scholes with Brent root-finding); runs on free data with no API key
- Walk-forward engine with strict t-1 data boundaries, sandboxed user strategies, and a parameter-sensitivity surface
- FastAPI backend and a React front end with 18+ panels, including a 3D implied-volatility surface
What it does not show
- Free data and a five-year evaluation window; the Sharpe comparison depends on including 2020.
Cite
BibTeX
@techreport{yadav2026voledge,
title = {VolEdge: Measuring and Trading the Variance Risk Premium with Model-Free Risk-Neutral Moments},
author = {Yadav, Tanishk},
year = {2026},
institution = {NYU Tandon School of Engineering},
type = {Working paper}
}
Tanishk Yadav
MS Financial Engineering at New York University, graduating May 2027, and looking for full-time quantitative research roles. My master's thesis measures, in real time, when US debt dynamics turn self-reinforcing.
More research
Other papers
RateWalk: Forecasting Fed Decisions
Naive inflation-regime conditioning overfits; empirical-Bayes shrinkage fixes it in the US, UK and Germany, and a blend with a market proxy reaches log-loss 1.056 and a 72.7% hit rate.
Calm Months, Crash Months
Built with only past data, capped leverage and trading costs, volatility management lifts momentum's Sharpe ratio from 0.43 to 0.94 and helps none of the other five factors. A closed-form formula for momentum's market beta explains why.
Multi-Scale Regime Detection
Four independent stress signals that must agree before de-risking: maximum drawdown -6.2% vs -41.7% for SPY, and the same calibration transfers to factor and international portfolios.