Hurst Exponent Package

Description

The function hurst takes a np.array of numbers and returns the Hurst exponent of the time series. The Hurst exponent is a measure of randomness of a time series. It is used in the study of long-term memory of time series. The value of the Hurst exponent is between 0 and 1. A value of 0.5 indicates that the time series is random. A value greater than 0.5 indicates that the time series is trending. A value less than 0.5 indicates that the time series is mean reverting.

Installation

pip install exp_hurst

Requirements

• numpy
• mmq

Usage

from coef_hurst import hurst
hurst(time_series)

Example

from exp_hurst import hurst
import numpy as np

# Create a time series of random numbers
rs = np.random.normal(0, 1, 100000)

# Evaluate the Hurst exponent
h = hurst(rs)

License

MIT

Author

[Igor Matheus Jasenovski]

Version

0.0.1

References

Hurst Exponent