Inferential Statisticshard
0:00.0
In a study using a saturated model, the Likelihood Ratio Test (LRT) statistic comparing a nested model with parameters to a model with parameters () is calculated. Under the null hypothesis that the additional parameters are zero, what is the geometric interpretation of in terms of the log-likelihood surface?
In a study using a saturated model, the Likelihood Ratio Test (LRT) statistic comparing a nested model with parameters to a model with parameters () is calculated. Under the null hypothesis that the additional parameters are zero, what is the geometric interpretation of in terms of the log-likelihood surface?