Full Strategy Research

This is the canonical QuantContext workflow: all three tools in sequence.

User prompt

“I want to build a value-momentum strategy. Screen for cheap stocks, rank by momentum, backtest it, and tell me if the alpha is real.”

Step 1:Preview the universe

Before backtesting, screen to see how many stocks pass the value filter today.

{
  "universe": "sp500",
  "screen_type": "fundamental_screen",
  "config": { "pe_lt": 15, "roe_gt": 10 }
}

Response: 28 stocks pass. Enough to build a diversified portfolio.

Step 2:Backtest the two-stage pipeline

Screen for value stocks (stage 1), then rank survivors by 200-day momentum and keep the top 30% (stage 2).

{
  "stages": [
    {
      "order": 1,
      "type": "screen",
      "skill": "fundamental_screen",
      "config": { "pe_lt": 15, "roe_gt": 10 }
    },
    {
      "order": 2,
      "type": "signal",
      "skill": "momentum_screen",
      "config": { "lookback_days": 200, "top_pct": 0.3 }
    }
  ],
  "universe": "sp500",
  "rebalance": "monthly",
  "sizing": "equal_weight",
  "start_date": "2022-01-01",
  "end_date": "2025-01-01",
  "max_position_size": 0.15,
  "stop_loss": 0.15
}

Response:

{
  "metrics": {
    "total_return": 0.4123,
    "cagr": 0.1216,
    "sharpe": 1.31,
    "max_drawdown": -0.1567,
    "calmar": 0.78,
    "win_rate": 0.5234,
    "turnover": 2.12,
    "total_trades": 256
  }
}

3-year performance: +41.2% total, 12.2% CAGR, Sharpe 1.31. Max drawdown held at -15.7%:the 15% stop-loss contributed. The strategy held 8–10 positions on average with 2x annual turnover.

Step 3:Factor analysis

{
  "equity_curve": ["...full_equity_curve from backtest..."]
}

Response:

{
  "alpha_annualized": 0.0534,
  "alpha_tstat": 1.92,
  "factors": {
    "Mkt-RF": { "loading": 0.7812, "tstat": 14.23 },
    "SMB":    { "loading": 0.1234, "tstat": 1.89 },
    "HML":    { "loading": 0.5678, "tstat": 7.34 },
    "Mom":    { "loading": 0.2345, "tstat": 3.45 }
  },
  "r_squared": 0.7456
}

Reading the results

Is the alpha real? Probably, but not conclusively.

Alpha is 5.3% annualized with t-stat 1.92:close to the significance threshold of 2.0 but just below it. With a longer backtest (5+ years), it might cross.
Strong value loading (HML 0.57):highly significant at t=7.34. The strategy is heavily tilted to the value factor. That’s the intended bet.
Moderate momentum loading (0.23):significant at t=3.45. The momentum ranking in stage 2 is contributing.
Market beta 0.78:defensive. This helped during the 2022 rate-hiking drawdown, which is why 3-year performance is solid despite 2022 headwinds.
R-squared 0.75:75% of variance explained by the four factors.

Bottom line: The strategy works, but most of the return comes from the value factor premium, not stock-selection alpha. You could replicate ~75% of the return with VLUE + MTUM ETFs. That’s not necessarily bad:the value premium is real and persistent:but it’s useful to know what you’re actually paying for.

Ways to increase the alpha component

Tighter value screen

Use pe_lt: 10 instead of pe_lt: 15 to target deeper value, which may have more idiosyncratic premium.

Add quality filter

Add roe_gt: 15 and debt_equity_lt: 0.5 before the momentum rank to avoid value traps.

Longer backtest

Extend to 5+ years for more data points and higher statistical power on the alpha t-stat.

Get Started

Tools

Examples

Reference

User prompt

Step 1:Preview the universe

Step 2:Backtest the two-stage pipeline

Step 3:Factor analysis

Reading the results

Ways to increase the alpha component

Tighter value screen

Add quality filter

Longer backtest

​User prompt

​Step 1:Preview the universe

​Step 2:Backtest the two-stage pipeline

​Step 3:Factor analysis

​Reading the results

​Ways to increase the alpha component

Tighter value screen

Add quality filter

Longer backtest

User prompt

Step 1:Preview the universe

Step 2:Backtest the two-stage pipeline

Step 3:Factor analysis

Reading the results

Ways to increase the alpha component