sigmaquant.research.autocorr.bartlett#

sigmaquant.research.autocorr.bartlett(values, lag, epsilon=0.05)#

Bartlett test for autocorrelation at a specific lag.

The test evaluates the null hypothesis that the autocorrelation at a given lag is equal to zero.

H0: rho_k = 0 H1: rho_k != 0

The variance of the sample autocorrelation is estimated using Bartlett’s formula:

Var(rho_k) ≈ (1 + 2 * sum_{j=1}^{k-1} rho_j^2) / n

The resulting Z-statistic is asymptotically standard normal.

Parameters:
  • values (ArrayLike) – Time series observations.

  • lag (int) – Lag at which the autocorrelation is tested.

  • epsilon (float, default 0.05) – Significance level used for the rejection decision.

Returns:

Structured container with the estimated autocorrelation, test statistic, p-value, and rejection decision.

Return type:

AutocorrelationTest