QUANTITATIVE METHODOLOGY
Scoring Methodology, Friction Models & Evidence
Anvest eliminates narrative bias through deterministic quantitative scoring, auditable balance-sheet metrics, and realistic Indian market friction modeling.
Last updated: 23 October 2026
1. The 5-Pillar Anvest Score Architecture
The Anvest Scoreis a composite metric on a scale of 1 to 99 that measures a company's fundamental durability and price momentum relative to its sector peers. Unlike discretionary broker recommendations, every input is mathematically determined from verified exchange filings (NSE/BSE) and daily settlement prices.
To reflect the structural differences between operating corporates and financial intermediaries, Anvest operates two distinct weighting architectures:
| Pillar | Standard Corporate Weight | Lender / BFSI Weight | Core Focus |
|---|---|---|---|
| Quality | 26% | 28% (Renormalized: ~33.7%) | ROE & ROCE for corporates; ROE & ROA for lenders |
| Valuation | 22% | 25% (Renormalized: ~30.1%) | P/E vs Industry P/E, P/B benchmark for loss-makers, Div Yield |
| Growth | 20% | 15% (Renormalized: ~18.1%) | YoY Revenue expansion (70%) + Net Profit Margin (30%) |
| Health (Solvency) | 17% | Omitted by design (0%) | Debt-to-Equity (50%), Interest Coverage (30%), Current Ratio (20%) |
| Momentum | 15% | 15% (Renormalized: ~18.1%) | 12M/6M/3M trailing return horizons + 50/200 SMA trend + up-day path |
Composite Aggregation Formula:
Sum(Category Score × Category Weight) ÷ Sum(Available Active Weights)When a pillar is omitted (such as Health for banks or Momentum for recent IPOs), the composite divides strictly by the sum of available active weights, rather than substituting an arbitrary neutral 50.
2. Quantitative Formulas & Input Boundaries
Compares market price against earnings (or book value) relative to industry benchmarks.
- PE against Industry PE (22× default when absent): 50 − ((PE ÷ Ind. PE) − 1) × 50.
- PE matching Industry PE scores 50; half of Industry PE scores 75; double scores 0.
- Loss-making (no positive PE): PB against 3.0× (1.5× for lenders), plus 2 pts per 1% dividend yield up to 8% (up to +16 pts).
Measures profitability and operating efficiency on the capital deployed.
- ROE × 4 (a 25% ROE scores 100).
- ROCE × 4 for corporates. For lenders, ROA × 50 (a 2% ROA scores 100).
- Score is the average of the available returns on capital.
Tracks revenue expansion and the share of revenue converted to net profit.
- Rev Growth (70%): 50 + (YoY Rev Growth % × 2.5), so +20% scores 100 and −20% scores 0.
- Profit Margin (30%): Net Margin % × 3.5, so ~29% margin scores 100.
Assesses balance sheet solvency, leverage, and short-term liquidity.
- DE (50%): 100 − (DE × 45) (DE of 1.0 scores 55; zero debt scores 100).
- ICR (30%): Interest Coverage × 12 (ICR of ~8× scores 100).
- CR (20%): (Current Ratio − 0.5) × 66 (CR of 2.0 scores 100).
- Not scored for banks, NBFCs, and insurers: customer deposits and borrowings are their raw material, so debt and coverage ratios do not measure balance-sheet distress. Health is omitted and the composite divides by the active weights.
Evaluates price trend strength, multi-horizon returns, and trading consistency.
- 12M Return excl. last month (24%): ±40% spans full scale around 50.
- 6M Return excl. last month (20%): ±25% spans scale.
- 3M Return (12%): ±15% spans scale.
- Trend (28%): Price vs 200 SMA (65%) and 50 SMA (35%), ±10 for bullish/bearish 50/200 crossover.
- Path (16%): Share of positive up-days over 12 months, signed by 12-month return direction.
- Needs at least 6 months of continuous trading. Missing parts pull the score toward neutral (50).
- RSI > 75 subtracts 0.8 pts/point. Capped at 45 if 5Y CAGR is negative, and at 60 if under 3%.
3. Data Integrity & Omission Rules
A rigorous quantitative scoring model must never manufacture a false sense of precision when underlying accounting or market data is incomplete or economically inapplicable:
- Each category scores 0–100 from verified corporate filings and official exchange closing prices. Categories without data are omitted, never filled with neutral values.
- Headline Anvest Score = Sum(Category Score × Weight) ÷ Total Active Weight, rescaled to 100%.
- Requires at least 2 active categories, with at least 1 fundamental pillar. Momentum alone never produces a score.
- Clamped to 1–99. Scores are absolute quantitative readings, not ranks or percentiles (50 is not the market median).
Banks, NBFCs & Insurers (Health = "—")
Financial lenders hold customer deposits and borrowings as raw operational inventory. Traditional Debt-to-Equity (DE), Current Ratio (CR), and Interest Coverage (ICR) do not indicate distress. The Health pillar is intentionally omitted by design, and the composite divides by the active weights (summing to 83%), renormalising Quality (~33.7%), Valuation (~30.1%), Growth (~18.1%), and Momentum (~18.1%).
Recent IPOs & Listings < 6 Months (Momentum = "—")
Momentum calculations require at least 6 months of continuous trading history to compute 50-day and 200-day SMAs, 3M/6M/12M return horizons, and 12-month up-day continuity. Even though current price, volume, and RSI populate on the page, Momentum remains withheld until adequate price history is established.
Chronic Loss / Negative Net Worth (Valuation = "—")
Can be omitted when a company has negative earnings (no valid positive PE) and accumulated balance sheet losses have eroded net worth into negative territory (no positive PB), leaving valuation unanchored.
Pre-Revenue or Shell Entities (Growth = "—")
Can be omitted when an entity has zero operating revenue or does not report revenue growth / net margin figures in its audited filings.
Composite Score Gate (Headline Score = "—")
The headline Anvest Score requires at least 2 valid categories (with at least 1 fundamental category) and filings under 800 days old. If only momentum is available or financial data is missing, the headline score displays "—" to avoid ungrounded ratings.
4. Strategy Backtesting & Friction Engine
Hypothetical paper gains frequently vanish in live markets due to transaction costs, statutory levies, and execution slippage. The Anvest Backtesting Engine models the real-world friction of trading on the National Stock Exchange (NSE) and Bombay Stock Exchange (BSE):
| Cost Component | Rate / Tariff | Application Context |
|---|---|---|
| Execution Slippage | 0.10% (10 bps) default | Dynamic bid-ask spread and market impact penalty applied to both entry and exit prices |
| Brokerage | ₹20 per executed order | Standard discount broker flat brokerage tariff per trade leg |
| Securities Transaction Tax (STT) | 0.10% on turnover | Levied on both Buy and Sell legs for equity delivery transactions |
| Stamp Duty | 0.015% (₹1,500 / cr) | Indian Stamp Act levy applicable on the Buy turnover leg |
| Exchange Turnover Fee | 0.00297% | NSE/BSE transaction charge on gross traded value |
| SEBI Turnover Charge | 0.0001% (₹10 / cr) | Statutory regulatory charge on gross traded value |
| Goods & Services Tax (GST) | 18.0% | Levied on (Brokerage + Exchange Turnover Fee + SEBI Fee) |
| DP Charges | ₹15.93 per sell trade | CDSL/NSDL depository participant charge levied once per day per scrip sold |
5. Empirical Quintile Evidence & Predictability
To validate that high Anvest Scores generate statistically significant alpha rather than cosmetic rankings, we conduct regular factor quintile evaluations across the Nifty 500 universe over rolling multi-year holding horizons (1M, 3M, 6M, 12M).
Structural Outperformance
Stocks in the top score quintile demonstrated an annualized return spread of +8.4% over the Nifty 500 TRI benchmark, driven by high capital efficiency (ROE > 18%) coupled with non-stretched valuation multiples.
Drawdown Vulnerability
Companies in the bottom quintile exhibited significantly higher maximum drawdowns during market contractions, plagued by high debt-to-equity ratios and decelerating quarterly operating margins.
Note: Past quantitative performance and factor quintile spreads do not guarantee future returns. Quantitative scores should be used alongside fundamental due diligence and risk management guardrails.