Hosmer-Lemeshow Test
To calculate the Hosmer-Lemeshow goodness of fit test, the data are sorted first in their increasing order of predicted probability. The observations are divided into groups using the setting for HL Groups, with a default of 10 groups. The groups are constructed based on the percentiles of the estimated probabilities. See Hosmer and Lemeshow (2000) for details.
The Hosmer-Lemeshow statistic is distributed as a
distribution with g-2 degrees of freedom given the null hypothesis.
The Hosmer-Lemeshow goodness of fit test statistic is given by:
