Log Return Node
Logarithmic Return — Per-Bar ln(P_t / P_{t-1})
Overview
The Log Return Node computes the natural logarithm of the price ratio between consecutive bars. Log returns are the standard representation used in quantitative finance because they are time-additive, symmetric, and approximately normally distributed for small returns.
They are the preferred input for risk metrics (CVaR, Sharpe, Hurst) and statistical tests that assume normally distributed data.
Formula
Inputs & Outputs
| Slot | Direction | Type | Description |
|---|---|---|---|
| input | Input | { values, timestamps } | Any positive numeric series (price, NAV, index) |
| values | Output | (number | null)[] | Log returns per bar; values[0] = null (1-bar warm-up) |
| timestamps | Output | number[] | Unix timestamps aligned to input |
Use Cases
Input to Risk Metrics
Feed log returns into CVaR, VaR, Sharpe, and other risk nodes for statistically sound results.
Strategy Return Analysis
Compute log returns of an equity curve to analyse the distributional properties of strategy performance.
Fractal Analysis Input
Log returns are the recommended input for Hurst Exponent, DFA, and other fractal/statistical indicators.
Tips & Best Practices
Log vs Simple Returns
Log returns are time-additive: sum of daily log returns = total log return. Use for multi-period aggregation. Simple returns are dollar-additive: use for portfolio weighting.
Positive Prices Required
Log return is undefined for zero or negative input. Ensure your price series is strictly positive before connecting to this node.
One Bar Warm-Up
The first output bar is always null because a prior bar is needed for the ratio. Account for this when computing downstream indicators.