tsqn-python 0.1.0

Robust time-series estimators based on the Qn scale estimator

tsqn (Python)

Python port of the original R package tsqn, providing robust time-series estimators based on the Qn scale estimator.

Features

• Robust covariance and correlation:
  • covQn, corQn
  • covMatQn, corMatQn
• Robust autocorrelation/autocovariance:
  • robacf
  • plot_robacf
• Robust periodograms:
  • PerQn
  • PerioMrob
• Long-memory estimation:
  • GPH_estimate

Install

pip install tsqn-python

Quick start

import numpy as np
from tsqn import covQn, corQn, robacf, GPH_estimate

rng = np.random.default_rng(42)
x = rng.normal(size=500)
y = 0.5 * x + rng.normal(size=500)

print(covQn(x, y))
print(corQn(x, y))

acf_out = robacf(np.column_stack([x, y]), lag_max=20, type="correlation", plot=False)
print(acf_out.acf.shape)  # (20, 2, 2)

print(GPH_estimate(x, method="GPH"))