c******5
发帖数: 22 | 1 In logistic regression,outcome is case (0 or 1).
Many covariates such as age, race ...Among those, I have two covariates
called bmi_2yrs(continous) and bmi_change_class (categorical)--- but bmi_
change (continous) is calculated by bmi_2yrs so we thought they should have
multicollinearity problem (bmi_change_class is generated by making the bmi_
change into different categories such as <5,5-10 and >10). Therefore, how
could I test multicollinearity for these covariates including age, race,
bmi_2 | c******5
发帖数: 22 | 2 没人理我呢？自己顶一下~~~ 实在是很着急，希望大家给点建议！多谢了！！ | s****n
发帖数: 620 | 3 我也是fresh啊，说错了莫怪。你可以将continuous variable划分区间变成
categorical variable，然后就可以用chi-square来检验两个categorical variable之
间的相关性。
have
the
a
【在 c******5 的大作中提到】
: In logistic regression,outcome is case (0 or 1).
: Many covariates such as age, race ...Among those, I have two covariates
: called bmi_2yrs(continous) and bmi_change_class (categorical)--- but bmi_
: change (continous) is calculated by bmi_2yrs so we thought they should have
: multicollinearity problem (bmi_change_class is generated by making the bmi_
: change into different categories such as <5,5-10 and >10). Therefore, how
: could I test multicollinearity for these covariates including age, race,
: bmi_2
| c******5
发帖数: 22 | 4 谢谢回答哈~但是我现在是在logistic regression 中modeling fitting时的问题：就
是说我想研究bim_change_class (categorical variable)与case 的关系，model中还
包含很多covariates（例如年龄，race什么的），还包括bmi_2yrs （这是个
continuous variable，把它变成categorical的就不一样了）， 是为了排除它带来的
因素。
你用chi-square来检验两个categorical variable之间的相关性没有错，但我其实不是
想知道bim_change_class和bmi_2yrs之间的关系。我是想知道在研究bmi_change_class
与outcome 的关系的model里，要是加进去了bmi_2yrs,会不会有什么multi-
collinearity的问题。我现在先做了一个只有bmi_change_class， 其他covariates（
例如年龄，race什么的）和outcome的model，发现adjusted for other covariat |
|
