> ## Documentation Index
> Fetch the complete documentation index at: https://quantcontext.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Momentum Strategy + Factor Decomposition

> Two-tool example:backtest a Nasdaq 100 momentum strategy and decompose the returns

This example uses `backtest_strategy` followed by `factor_analysis`. It's the most common two-step workflow: test a strategy, then understand where the returns came from.

## User prompt

> "I want to test a strategy that buys the top 20% momentum stocks in the Nasdaq 100, rebalanced monthly. How would it have done over the last 2 years?"

## Step 1:Backtest

```json theme={null}
{
  "stages": [
    {
      "order": 1,
      "type": "screen",
      "skill": "momentum_screen",
      "config": { "lookback_days": 200, "top_pct": 0.2 }
    }
  ],
  "universe": "nasdaq100",
  "rebalance": "monthly",
  "sizing": "equal_weight",
  "start_date": "2023-03-01",
  "end_date": "2025-03-01"
}
```

**Response:**

```json theme={null}
{
  "metrics": {
    "total_return": 0.4812,
    "cagr": 0.2187,
    "sharpe": 1.56,
    "max_drawdown": -0.1234,
    "calmar": 1.77,
    "win_rate": 0.5523,
    "turnover": 3.82,
    "total_trades": 412
  },
  "equity_curve_points": 504,
  "full_equity_curve": ["...504 points, pass to factor_analysis..."]
}
```

**Agent summary:**

* **Total return:** +48.1% over 2 years (21.9% CAGR)
* **Sharpe ratio:** 1.56:strong risk-adjusted returns
* **Max drawdown:** -12.3%:manageable
* **Calmar ratio:** 1.77:good return per unit of peak-to-trough risk
* **Turnover:** 3.82x/year:fairly active at 412 trades (expected for momentum)

## Step 2:Factor decomposition

The user follows up: *"Where is the return coming from?"*

```json theme={null}
{
  "equity_curve": ["...full_equity_curve from backtest..."]
}
```

**Response:**

```json theme={null}
{
  "alpha_daily": 0.000312,
  "alpha_annualized": 0.0821,
  "alpha_tstat": 2.34,
  "factors": {
    "Mkt-RF": { "loading": 1.1234, "tstat": 18.92 },
    "SMB":    { "loading": -0.3456, "tstat": -4.12 },
    "HML":    { "loading": -0.2891, "tstat": -3.45 },
    "Mom":    { "loading": 0.4523,  "tstat": 6.78 }
  },
  "r_squared": 0.8234,
  "residual_vol": 0.0912,
  "interpretation": {
    "alpha": "Alpha is statistically significant (t-stat=2.34)",
    "factors": "Dominant factor: Mkt-RF (loading=1.1234)"
  }
}
```

**Agent interpretation:**

**Alpha is statistically significant**:8.2% annualized, t-stat 2.34. There is genuine edge beyond factor exposure.

**Factor breakdown:**

* **Market (1.12)**:Slightly higher beta than the market. Amplifies both gains and losses.
* **Size (-0.35)**:Large-cap tilt, expected for Nasdaq 100.
* **Value (-0.29)**:Anti-value tilt, expected for a momentum strategy that favors expensive growth stocks.
* **Momentum (0.45)**:Significant momentum loading, as expected. But the stock selection within the momentum bucket is adding extra value beyond the factor alone.

**R-squared: 0.82**:82% of variance explained by factors. The remaining 18% is idiosyncratic.

<Warning>
  The 1.12 market beta means this strategy will significantly underperform in a bear market. The strong backtest period (2023–2025) was largely bullish:consider stress-testing on 2022 data.
</Warning>