Andrews.test {CPAT}

Andrews' Test for End-of-Sample Structural Change

Description

Performs Andrews' test for end-of-sample structural change, as described in (Andrews 2003). This function works for both univariate and multivariate data depending on the nature of x and whether formula is specified. This function is thus an interface to andrews_test and andrews_test_reg; see the documentation of those functions for more details.

Usage

Andrews.test(x, M, formula = NULL)

Arguments

x

Data to test for change in mean (either a vector or data.frame)

M

Numeric index of the location of the first potential change point

formula

The regression formula, which will be passed to lm

Value

A htest-class object containing the results of the test

References

Andrews DWK (2003). “End-of-Sample Instability Tests.” Econometrica, 71(6), 1661–1694. ISSN 00129682, 14680262, https://www.jstor.org/stable/1555535.

Examples

Andrews.test(rnorm(1000), M = 900)
x <- rnorm(1000)
y <- 1 + 2 * x + rnorm(1000)
df <- data.frame(x, y)
Andrews.test(df, y ~ x, M = 900)