dx_g_test: G-Test (Log-Likelihood Ratio Test) for Independence in 2x2...

In overdodactyl/diagnosticSummary: Diagnostic Summaries

G-Test (Log-Likelihood Ratio Test) for Independence in 2x2 Table

Description

Conducts a G-test of independence to assess the goodness of fit or the association between two categorical variables in a 2x2 contingency table. It is an alternative to the chi-squared test and is particularly useful when dealing with small expected frequencies.

Usage

dx_g_test(cm, detail = "full")

Arguments

cm	A dx_cm object created by dx_cm().
detail	Character specifying the level of detail in the output: "simple" for raw estimate, "full" for detailed estimate including 95% confidence intervals.

Details

The test compares the observed frequencies to the expected frequencies based on the marginal totals and calculates a G statistic, which follows a chi-squared distribution. The test is especially useful when the data contains small expected frequencies, which might make the chi-squared test less accurate. A low p-value indicates a significant association between the variables or a significant difference from the expected distribution. Caution is needed with zero counts or very small samples.

Examples

cm <- dx_cm(dx_heart_failure$predicted, dx_heart_failure$truth,
  threshold = 0.3, poslabel = 1
)
simple <- dx_g_test(cm, detail = "simple")
detailed <- dx_g_test(cm)
print(simple)
print(detailed)

overdodactyl/diagnosticSummary documentation built on Jan. 28, 2024, 10:07 a.m.