I am trying to calculate LR_statistic for two models where my log-likelihood ratio values as follows.
model1.llf = -2326.9 and model2.llf = -4649.1.
When read online some of the articles showing The LR statistic is calculated by taking the difference between the log-likelihood values of the two models and multiplying it by -2 or +2.
LR_statistic = -2 * (model1.llf - model2.llf) = -4644.29 , p_val = 1.0
LR_statistic = 2 * (model1.llf - model2.llf) = 4644.29, p_val = 0.0
Can anyone explain whether I have to multiple by -2 or +2 and explain the reason.
