Hurst Exponent Pass Node

Hurst Exponent — Series Input

StatisticalFractalPass

Overview

The Hurst Exponent Pass Node applies Rescaled Range (R/S) analysis to any upstream numeric series, computing the Hurst exponent H over a rolling window. H measures the long-range dependence or memory of the series.

The Hurst exponent was originally used in hydrology to characterise river flood cycles and has become a key tool in quantitative finance for distinguishing trending from mean-reverting markets.

Formula

Rescaled Range (R/S) Analysis:
R = max(cumulative deviation) − min(cumulative deviation)
S = standard deviation of the window
H = log(R/S) / log(n/2)
H > 0.5 = persistent (trending); H < 0.5 = anti-persistent (mean-reverting); H = 0.5 = random walk

Parameters

ParameterDefaultDescription
period100Rolling window in bars. Values ≥ 100 recommended for reliable estimates.

Inputs & Outputs

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

Use Cases

Strategy Selection

H > 0.55 → trend-following; H < 0.45 → mean-reversion; 0.45–0.55 → neutral / wait for clarity.

Market Regime Classification

Track H over time to classify bull trends, consolidations, and mean-reverting corrective phases.

Asset Screening

Screen assets for persistent (H > 0.6) behaviour to find the best candidates for momentum strategies.

Tips & Best Practices

Use Large Windows

R/S analysis is noisy with small samples. Use period ≥ 100 for daily data; ≥ 200 for even more reliable estimates.

Log Returns Input

Feed log returns to the Hurst node for a more stationary input that produces cleaner estimates than raw prices.

Confirm with DFA

DFA is more robust to non-stationarity. Use both Hurst and DFA to confirm the regime before trading.

Related Indicators