Rolling Standard Error Pass Node
Rolling Standard Error of the Mean — Series Input
Overview
The Rolling Standard Error Pass Node computes the Standard Error of the Mean (SEM) over a rolling window — a measure of the uncertainty around the rolling mean estimate.
SEM is smaller than standard deviation by a factor of √n. It represents the likely range of error in the rolling mean estimate. When SEM is small relative to the mean, the mean is a reliable estimate; when SEM is large, the mean is uncertain.
Formula
Parameters
| Parameter | Default | Description |
|---|---|---|
| period | 20 | Rolling window in bars |
Inputs & Outputs
| Slot | Direction | Type | Description |
|---|---|---|---|
| input | Input | { values, timestamps } | Any upstream numeric series |
| values | Output | (number | null)[] | Standard error of the mean per bar; nulls during warm-up |
| timestamps | Output | number[] | Unix timestamps aligned to input |
Use Cases
Signal Confidence Bands
Add and subtract 1.96×SE from the rolling mean to construct 95% confidence intervals — trade only when the signal exceeds the confidence band.
Estimation Uncertainty
Monitor rolling SEM to assess how reliable mean-based signals are — high SEM = unreliable mean estimates, suggesting noise-dominated regime.
IC Information Ratio
Divide Rolling IC by Rolling Standard Error of IC to compute the IC Information Ratio — measures consistency of predictive power.
Tips & Best Practices
SE vs StdDev
SE = σ/√n is always smaller than std deviation σ. Use std for volatility; use SE for uncertainty around the mean estimate itself.
Larger Windows Reduce SE
SE decreases as 1/√period — doubling the window halves the SE. Use longer periods when you need more precise mean estimates.
Used in Hypothesis Testing
SE is the denominator in t-tests. Use it to test whether the rolling mean return is statistically different from zero: t = mean/SE.