s***i
发帖数: 49 | 1 如果一个logit regression, 没有explanatory variable (除了 constant term), 它
的deviance 是不是 0 ？
有个logit regression, 2个explanatory variables, A and B. 我现在要做一个drop-
in- deviance test，hypothesis 是 "neither A nor B" is significant。 能不能找
到full model 的 deviance, 然后reduced model 就是A 和 B 都没有，什么都没有，
就是deviance=0. 所以这个 drop in deviance 就是 full model的deviance.
对吗？ |
g*********n
发帖数: 119 | 2 the deviance is a G^2, compared with saturated model, so a logistic
regression with only intercept will have the largest deviance(>0), instead
of 0.
the deviance of reduced model shall >= the deviance of full model, so I
believe what you think is not true. |
s***i
发帖数: 49 | 3 So how can I answer the question? I already know the deviance of the full
model, but I do not know the deviance of the reduced model (with only
intercept, because it asks to test the question "neither A nor B is an
important variable" using drop-in-deviance test). So what is the drop in
deviance here?
【在 g*********n 的大作中提到】
: the deviance is a G^2, compared with saturated model, so a logistic
: regression with only intercept will have the largest deviance(>0), instead
: of 0.
: the deviance of reduced model shall >= the deviance of full model, so I
: believe what you think is not true.
|
h******a
发帖数: 198 | 4 For the test statistics should be
-2(Loglikehood reduced model - Loglikelihood full model) follows Chi-squre
SAS会帮你算reduced model的likelihood的啊 |
g*********n
发帖数: 119 | 5 you can use R with GLM to achieve respective deviance for two models, then
use the chi square test with df=2 to test the hypothesis. |
s***i
发帖数: 49 | 6 Can I use the "null deviance" in R as the deviance with 0 explanatory
variable? But how do I model a logit regression with 0 explanatory variable
in the first place, using GLM? |
s***i
发帖数: 49 | 7 Hmm ok. Although I am using R. |
g*********n
发帖数: 119 | 8 can I use the "null deviance" in R as the deviance with 0 explanatory
variable? |
s***i
发帖数: 49 | 9 So "null deviance" of a full model is the same as any of the models with
some variables dropped? |
