Chow Test

This project provides an implementation of the Chow break test.

The Chow test was initially developed by Gregory Chow in 1960 to test whether one regression or two or more regressions best characterise the data. As such, the Chow test is capable of detecting "structural breaks" within time-series. Additional information can be obtained from:

Chow, Gregory C. "Tests of equality between sets of coefficients in two linear regressions." Econometrica: Journal of the Econometric Society (1960): 591-605.

Toyoda, Toshihisa. "Use of the Chow test under heteroscedasticity." Econometrica: Journal of the Econometric Society (1974): 601-608.

This implementation supports simple linear models, and finding breaks where k = 2.

Installation

This module requires Python 3.0+ to run. The module can can be imported by:

pip install chowtest
from chow_test import chow_test

Input Arguments

The required input arguments are listed below:

Argument	Description	Data Type
X_series	The series or series' denoting the X variables.	pd.Series or pd.DataFrame
y_series	The series denoting the y variable	pd.Series
last_index	The final index value to be included before the data split.	int
first_index	The first index value to be included after the index split.	int
significance	The significance level against which the p-value is assessed. Can be 0.1, 0.05 or 0.01	float

Note: The pd.reset_index() function can be invoked in order to obtain an integer index.