Return Skewness

Quick Reference

Property	Value
Dimension	regime
Category	volatility
Version	v0.9.0 (Beta)
Output Column	skewness

Return skewness: rolling_skew(ret, window) - asymmetry in ret distribution

Formula

rolling_skew(ret, window)

CDM Inputs

Column	CDM Table	Description
ret	cdm_trade_enriched	Trade data enriched with bar context 鈥?price, volume, side, trade type

Parameters

Parameter	Type	Default	Description
window	integer [5, 1000]	50	Window size for skewness calculation

Output

Column: skewness

Rolling skewness of returns

Market Intuition & Trading Rationale

Return skewness measures the asymmetry of the return distribution: rolling_skew(ret, window). Positive skew means the distribution has a long right tail 鈥?large positive returns are more common than large negative returns. Negative skew means a long left tail 鈥?large negative returns dominate. Zero skew means the distribution is symmetric (like a normal distribution).

Equity markets typically exhibit negative skew (crashes are larger and faster than rallies). Crypto markets can exhibit either positive or negative skew depending on the regime 鈥?bull markets often show positive skew (explosive upside), while bear markets show extreme negative skew (cascading liquidations). The rolling nature of this feature captures these regime-dependent skew changes.

Skew is a third-moment statistic and requires more data than mean (first moment) or variance (second moment) for reliable estimation. The window should be at least 30 observations for meaningful skew estimates, and 100+ for stable ones. Short-window skew estimates are noisy and should be interpreted cautiously.

Usage Cases

• Tail risk assessment: Strongly negative skew 鈫?the instrument is prone to crashes. Position sizing should account for the possibility of extreme negative returns beyond what variance alone would suggest.
• Options strategy selection: Negative skew favors put-selling strategies (the market overprices crash protection). Positive skew favors call-selling strategies. The skew_regime determines which options strategy is statistically favored.
• regime context: Used in intraday_momentum pack 鈥?return_skewness provides regime context for momentum signals. Positive skew + positive momentum = trend is likely to continue with explosive upside. Negative skew + positive momentum = cautious 鈥?the trend may end in a crash.

YAML Definition

name: return_skewness
description: 'Return skewness: rolling_skew(ret, window) - asymmetry in ret distribution'
category: volatility
version: v0.9.0 (Beta)
dimension: regime
status: Pre-release
required_inputs:
  - ret
output_column: skewness
output_description: Rolling skewness of returns
parameters:
  window:
    type: integer
    description: Window size for skewness calculation
    required: false
    default: 50
    constraints:
      min: 5
      max: 1000
formula: rolling_skew(ret, window)