我有以下公式:
glm1 <- glm(y ~ g1 + g2, data = fish, family = binomial)
y是二项式变量: num 1 1 0 0 1 0 0 0 0 1 ... #有 (1) 或没有 (0) 鱼。 g1:因子 w/ 4 个级别“1”、“2”、“3”、“4”:1 2 1 1 4 1 2 1 1 3 ... #River 级别。 g2 : num 808 1110 754 421 568 535 570 690 668 556 ... #降水量 (mL)。
length(y)
>150
length(g1)
>150
length(g2)
>150
为了测试同方差性,我尝试了:
bartlett.test(split(y, list(g1,g2))) #Then I get this:
>Error in bartlett.test.default(split(y, list(g1, :
> there must be at least 2 observations in each group. #I think it's strange, because I've 3 vectors at the formula.
但是我的公式是y~g1+g2,那么:
bartlett.test(y ~ g1+g2, data=fish) #Then I get this:
>Error in fligner.test.formula(y ~ g1 + :
> 'formula' should be of the form response ~ group
我还尝试了其他测试:
leveneTest(y ~ g1+g2) #Then I get this:
>Error in leveneTest.formula(y ~ g1 + g2, data = fish) :
> Levene's test is not appropriate with quantitative explanatory variables.
最后,我尝试了这个:
fligner.test(split(y, list(g1,g2)))
>Error in fligner.test.default(split(y, list(g1, :
> all groups must contain data #Again, I think it's strange, because I've 3 vectors at the formula
1. g1 和 g2 都必须是因子(分类变量)来测试同方差性吗?2. 如何用 Bartlett's、Levene's 和 Fligner's 在我的公式(y-num ~ g1-factor + g2-num)中测试同方差性?
我提前感谢您的帮助。