Rolling Skewness Pass Node

Rolling Skewness — Series Input

StatisticalDistributionalPass

Overview

The Rolling Skewness Pass Node computes the rolling skewness of a series — the third standardised moment, measuring the asymmetry of the return distribution within a sliding window.

Positive skewness (right tail) means occasional large gains; negative skewness (left tail) means occasional large losses. Financial return distributions typically exhibit negative skewness.

Formula

Rolling Skewness:
skew[i] = [Σ(xⱼ − μ)³ / n] / σ³
μ = rolling mean, σ = rolling standard deviation, over window of size n=period.
skew > 0 = right-skewed (long right tail). skew < 0 = left-skewed (long left tail). skew = 0 = symmetric.

Parameters

ParameterDefaultDescription
period20Rolling window in bars

Inputs & Outputs

SlotDirectionTypeDescription
inputInput{ values, timestamps }Any upstream numeric series
valuesOutput(number | null)[]Rolling skewness per bar; nulls during warm-up
timestampsOutputnumber[]Unix timestamps aligned to input

Use Cases

Tail Risk Direction

Negative skewness in returns signals greater downside tail risk — consider hedging or reducing leverage during periods of persistently negative rolling skewness.

Options Skew Analysis

Negative rolling skewness on return series corresponds to put options being more expensive than calls — useful for volatility surface analysis.

Distribution Shape Profiling

Combine with kurtosis for a full distributional picture: skewness gives direction of asymmetry; kurtosis gives magnitude of tail weight.

Tips & Best Practices

Apply to Returns, Not Prices

Skewness of raw price levels is dominated by trend effects. Feed log-returns or simple returns for meaningful distributional analysis.

Needs Sufficient Sample

Skewness estimates are unreliable for period < 20. Use period ≥ 30 for stable estimates; skewness is typically the noisiest of the moments.

Pair with Kurtosis

Skewness and kurtosis together characterise deviation from normality. High negative skewness + high kurtosis = crash-prone, fat left tail.

Related Indicators