Concept

Autocorrelation Periodogram

Autocorrelation Periodogram is a Statistics concept. The Library holds 1 implementations, each one a working definition you can pull into Quant.

Top Autocorrelation Periodogram indicators

1 total

What is the Autocorrelation Periodogram?

The autocorrelation periodogram is John Ehlers' spectral estimator for finding the dominant cycle in price. Instead of Fourier-transforming price directly, it first computes the autocorrelation of a band-limited version of price at a range of lags, then measures the power at each candidate period by correlating that autocorrelation series against sine and cosine waves of that period. The theoretical license is the Wiener-Khinchin relation: the power spectrum of a stationary series is the Fourier transform of its autocorrelation function. Working through autocorrelation adds a practical benefit: correlation values are bounded between -1 and +1, so the spectrum is normalized for swing size and quiet and volatile stretches become comparable.

In Ehlers' presentation the input is conditioned with a roofing filter to strip trend and high-frequency noise, the scan covers a bounded band of tradeable periods (roughly 10 to 48 bars in his examples), and the result is drawn as a spectrogram heatmap: time across, period up, color for normalized power. The dominant cycle is read as the power-weighted center of the strong periods and typically feeds dominant cycle measurement workflows that adapt indicator lookbacks. Market cycles drift, fade, and sometimes vanish entirely, so a smeared or shifting ridge is itself information: it says there is no stable cycle to tune to.

How traders use it

  • Adaptive lookbacks: the measured dominant cycle sets oscillator and moving-average lengths on the fly (for example, tuning a stochastic to a fraction of the dominant period), replacing one fixed setting with a cycle-aware one.
  • Regime filtering: a sharp, persistent ridge on the spectrogram argues for cycle-based tactics such as fading swings, while weak or smeared power argues the market is trending or noisy and cycle tools should stand down.
  • Auditing cycle claims: because the display shows the whole spectrum through time, it exposes whether an apparent cycle was stable across history or an artifact of one lucky stretch.

Related concepts · Ehlers DSP cluster

Concept family

Statistics

45 concepts mapped · 37 in the Library

Autocorrelation Periodogram FAQ

How is the autocorrelation periodogram different from a plain FFT of price?

Mathematically they target the same spectrum: the Wiener-Khinchin theorem links the two. Practically, transforming the autocorrelation instead of raw price normalizes amplitude, because correlations are bounded between -1 and +1, and Ehlers argues it resolves short data windows better. A plain FFT applied to price needs detrending, windowing, and relatively long stationary samples, which market data rarely supplies.

What does the dominant cycle reading actually tell you?

It estimates the period, in bars, currently carrying the most spectral power, which is then used to tune lookbacks and anticipate swing rhythm. It is an estimate, not a schedule: market cycles drift and disappear, and during strong trends there may be no meaningful cycle at all. Most workflows require a clear, persistent spectral ridge before letting the reading drive parameters.

Build Autocorrelation Periodogram your way.

Quant writes, tests, and refines it with you — then it runs on LuxAlgo charting or ports to TradingView.