Name: Tail Mean Difference (TMD)
Brand: AlpineDataWorks.AI
SKU: ADW-106
Availability: InStock

Use cases

What it unlocks

⌁

For an agent

A factor-rotation agent reads ADW-106 each day; when the TMD score rises above 55 (50 = symmetric; current is 56.8, 77th percentile of a 2,454-day history with trend rising) and upper_tail_mean exceeds |lower_tail_mean|, it tilts the factor allocation toward momentum and growth names that benefit from positive skew and de-weights low-volatility defensive factors, embedding the IOM's tmd_raw and methodology_version in the rebalancing log for attribution analysis. A score dropping back below 48 — signaling downside skew — triggers the reverse tilt automatically.

📈

For the business

A quantitative portfolio manager uses ADW-106's upper_tail_mean and lower_tail_mean fields to inform options-skew trading: when TMD is above 55, right-tail upside is fatter than the left tail, making put spreads relatively cheap and call spreads relatively expensive on a distribution-adjusted basis, which the PM uses to justify a skew-selling strategy. Unlike standard skewness statistics pulled from raw returns, TMD's 60-day normalized output arrives pre-packaged in the IOM structure with source lineage and confidence, saving the quant team the step of cleaning and aligning OHLCV data before running their own tail-mean computation.

Forward outlook

Prediction

Horizon: —
Recommended use: Capture asymmetric return skew. Score > 55 = upside-dominant tail; < 45 = downside-dominant — adjust directional bias accordingly.

Methodology

How it's built

60-day returns → μ_up (90th pct tail) − |μ_low| (10th pct tail) → normalize by |mean| → map to 0-100 (50=symmetric, >50=upside skew)

Stooq daily OHLCV (SPY, keyless)

Version ADW-106-live-1.0 · validated to beat a naive baseline · benchmark: none

Live response

The object an agent receives

One call returns the answer with its reasoning attached — the live Intelligence Object for ADW-106.

GET /v1/intelligence/adw-106

{
  "product_id": "ADW-106",
  "entity": "S&P 500 (SPY)",
  "score": 56.8,
  "trend": "symmetric",
  "confidence": 0.78,
  "top_drivers": [
    {
      "factor": "tmd_raw",
      "contribution": 0.002655
    },
    {
      "factor": "upper_tail_mean",
      "contribution": 0.016236
    },
    {
      "factor": "lower_tail_mean_abs",
      "contribution": 0.01358
    }
  ],
  "recommended_use": "Capture asymmetric return skew. Score > 55 = upside-dominant tail; < 45 = downside-dominant — adjust directional bias accordingly.",
  "methodology_version": "ADW-106-live-1.0",
  "freshness": "2026-06-26T20:00:16.189Z",
  "coverage": "S&P 500 ETF daily returns — 60-day rolling window",
  "source_lineage": [
    "Stooq daily OHLCV (SPY)"
  ],
  "allowed_use": "informational",
  "tmd_raw": 0.002655,
  "tmd_normalized": 0.1352,
  "upper_tail_mean": 0.016236,
  "lower_tail_mean": -0.01358,
  "upper_tail_pct": 0.9,
  "lower_tail_pct": 0.1,
  "window_days": 60,
  "skew_label": "symmetric",
  "validation_status": "descriptive"
}

IOM schema