Introductory Context
"Quantitative multi-asset trend following strategy using liquid Indian ETFs. Featuring causal formation returns, moving average gating, and inverse-variance portfolio optimization."
Liquid Multi-Asset Trend Following: The Long-Only ETF Portfolio
Institutional managed-futures trend following (CTA style) typically uses cross-asset derivative futures to go both long and short across dozens of global markets. However, many family offices, wealth managers, and retail quantitative investors in India operate under long-only mandates or wish to avoid the operational complexity, margin calls, and quarterly rollover costs of futures contracts.
Multi-Asset Trend Following with ETFs provides an elegant, highly liquid alternative. The strategy systematically evaluates a diversified panel of liquid ETFs spanning equities, fixed income, precious metals, and international markets. At every monthly rebalance, only assets exhibiting positive trailing momentum (R_cum > 0) and trading above their 200-day moving average are retained. Rejected assets with negative momentum are not shorted; instead, their capital is converted to risk-free sovereign cash (Liquid BeES / T-Bills).
• Strategy Logic at a Glance
For a multi-asset universe of liquid Indian ETFs (Large-cap Equities, Mid-caps, Bank, Gold, Silver, Long-term Gilts, and International Tech), compute the trailing 12-month return R_cum. Retain only ETFs with R_cum > 0 and Price > 200 DMA. Weight survivors using Inverse-Variance Scaling (w_i ∝ R_cum_i / σ_i²) capped at 25% per asset, allocating all remaining capital into Liquid BeES / Overnight Cash.
Mathematical Formulation & Weighting Schemes
For each ETF i in the universe, the 12-month simple formation return is measured across window [t1, t2] strictly ending prior to trade execution:
On the survivor set (where R_cum_i > 0), quantitative portfolio managers can select between three weighting methodologies:
- Scheme 1 (Pure Momentum): w_i = γ1 × R_cum_i. Capital is proportional to raw performance, but this tends to over-allocate to volatile equity ETFs.
- Scheme 2 (Risk Parity Momentum): w_i = γ2 × (R_cum_i / σ_i). Standardizes returns by realized annualized volatility over the trailing 60 to 126 trading days.
- Scheme 3 (Inverse-Variance Sharpe Optimizer - Default): w_i = γ3 × (R_cum_i / σ_i²). Maximizes portfolio Sharpe under a diagonal covariance assumption, heavily penalizing high-volatility spikes.
Normalizing constants γ are computed such that survivor weights sum to 1. An individual position cap w_max = 25% prevents any single runaway commodity or equity ETF from dominating the risk budget.
Indian Multi-Asset Universe on the National Stock Exchange (NSE)
The strategy is deployed across liquid Indian exchange-traded funds representing non-correlated macroeconomic asset classes:
| Asset Class | NSE ETF Ticker | Underlying Index / Asset | Macroeconomic Role |
|---|---|---|---|
| Large-Cap Equities | NIFTYBEES | Nifty 50 Index | Core Indian economic growth engine |
| Mid-Cap Equities | MID150BEES | Nifty Midcap 150 Index | High-beta domestic expansion |
| Precious Metals (Gold) | GOLDBEES | Domestic Spot Gold (INR) | Currency hedge & geopolitical crisis safe-haven |
| Precious Metals (Silver) | SILVERBEES | Domestic Spot Silver (INR) | Industrial recovery & monetary inflation hedge |
| Long-Term Sovereign Debt | GILT5YBEES / LTGILTBEES | Indian 8-13 Year Sovereign G-Secs | Deflationary hedge & RBI rate-cut beneficiary |
| International Equities | MON100 / MASPTOP50 | NASDAQ 100 / S&P 500 Index | Global tech innovation & USD/INR depreciation |
| Cash Parking Residual | LIQUIDBEES | TREPS / Tri-Party Overnight Repo | Zero-risk parking lot for rejected assets |
The Rebalancing and Execution Protocol
Multi-asset ETF trend portfolios follow strict systematic operating guidelines:
- Delay-1 Rebalancing: Formation return R_cum and volatility σ are computed using finalized month-end closing prices. Orders are executed at the market open on the first trading session of the new calendar month (never at the same closing print used to compute signals).
- Cash Parking vs. Cash Drag: When macroeconomic turmoil forces all equity, bond, and commodity ETFs into negative territory (such as late 2008 or early 2020), 100% of the portfolio shifts into Liquid BeES. Backtests must explicitly mark interest yields on cash rather than assuming a 0% return.
- Tracking Error Awareness: Indian commodity ETFs (Gold and Silver) reflect domestic landed prices, import duties, and tracking error versus international futures. Using international spot prices in backtests creates unrealistic model alpha.
Risk Factor: Diagonal Covariance & Crisis Co-Movement
During liquidity shocks and black swan events, historical asset correlations can collapse toward +1.0 as market participants liquidate all holdings simultaneously. Inverse-variance weighting provides robust risk budgeting during standard regimes, but does not prevent short-term drawdowns when multiple asset classes experience sudden, simultaneous down-gaps.
Educational and Research Content Only
This strategy specification is provided strictly for educational, backtesting, and quantitative research purposes. ETF prices fluctuate with market conditions, tracking error, and liquidity. It does not constitute investment advice or a recommendation to purchase securities.