The Generalized Linear Model computes several assessment measures to help you evaluate
how well your model fits the data. These assessment measures are available at the
top of the model pane. Click the currently displayed
assessment measure to see all available assessment measures. The available assessment measures are the
following:
-2 Log Likelihood
The
likelihood function estimates the probability of an observed sample given all possible
parameter values. The
log likelihood is simply the logarithm of the likelihood function. This value is -2 times the log
likelihood. Smaller values are preferred.
AIC
Akaike’s Information Criterion. Smaller values indicate better models.
AIC values should be compared only when two models have an approximately equal number
of observations. AIC values can become negative. AIC is based on the Kullback-Leibler
information measure of discrepancy between the true distribution of the
response variable and the distribution specified by the model.
AICC
Corrected Akaike’s Information Criterion. This version of AIC adjusts the value to
account for a relatively small
sample size. The result is that extra effects penalize AICC more than AIC. As the sample size
increases, AICC and AIC converge.
BIC
The Bayesian Information Criterion (BIC), also known as Schwarz’s Bayesian Criterion
(SBC), is an increasing function of the model’s
residual sum of squares and the number of effects. Unexplained variations in the response variable and the
number of effects increase the value of the BIC. As a result, a lower BIC
implies either fewer explanatory variables, better fit, or both. BIC penalizes free
parameters more strongly than AIC.