# What is the likelihood ratio for this chi-square test (assocstats)

I am using the assocstats function in the vcd library. It gives something called the likelihood ratio and its associated p-value which is different from the chi-squared statistic and its p-value. What is this likelihood ratio and what is it's distribution?

e.g.,

> library(vcd)
> data("Arthritis")
> tab <- xtabs(~Improved + Treatment, data = Arthritis)
> assocstats(tab)
X^2 df  P(> X^2)
Likelihood Ratio 13.530  2 0.0011536
Pearson          13.055  2 0.0014626

Phi-Coefficient   : NA
Contingency Coeff.: 0.367
Cramer's V        : 0.394
> chisq.test(tab)

Pearson's Chi-squared test

data:  tab
X-squared = 13.055, df = 2, p-value = 0.001463


A test using that likelihood ratio is generally called G-test. With r×s contingency table, $G = 2 \sum_{i=1}^r \sum_{j=1}^s O_{ij} * ln (O_{ij}/E_{ij})$. $O_{ij}$ is the observed count and $E_{ij}$ is the expected count under the null hypothesis. Distribution of G is approximately a chi-squared distribution. G-test (wikipedia) gives more information.

observe <- mar_table(tab)[1:3, 1:2]  # delete total
expect <- independence_table(tab)

# (observe) Treatment          (expect) Treatment
# Improved Placebo Treated    Improved   Placebo   Treated
#   None        29      13       None   21.500000 20.500000
#   Some         7       7       Some    7.166667  6.833333
#   Marked       7      21       Marked 14.333333 13.666667

G <- 2 * sum( observe * log(observe/expect) )  # [1] 13.52981
chisq <- sum( (observe - expect)^2/expect )      # [1] 13.05502 # as a reference

1 - pchisq(G, df = 2)             # [1] 0.001153559
1 - pchisq(chisq, df = 2)           # [1] 0.001462643