用R做的方差分析结果，怎么解读？

相似文件 换一批

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

这是我用这是我用R做的y对x1,x2,x3,x4的多元回归的检验结果。有两个问题想请教一下：①“Sum Sq”和"Mean Sq"那两列的数值为什么是一样的？②为什么x1,x2,x3,x4都有一个F值。按我的理解，要做F检验的话不是只有一个F值吗？而且应该是针对y计算的F值呀？

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：方差分析 mean 多元回归 F检验 Sum 而且

第一个问题那两列一样是因为自由度为1，sum/1=Mean
第二个问题多元回归除对回归方程要做显著性检验外还需要需要对回归系数做显著性检验，这就是会有F的原因，当然这里也可以用t检验，但是可以用t检验那就一定可以用F检验，这个从统计量就能看出来
回答完毕，希望对lz有用

In response to question 2, You should know that you are fitting a ordinary linear model with 4 predictors (not the traditional one way ANOVA for single factor problem), that is why you got 4 F values. Also take care when you are trying to interpret them, since the variables are not mutually orthogonal. (check the concepts such as "sequential sum squares" and "partial sum squares")
To sum, before applying any software to data, try to understand what model you used and how to interpret it.

hugebear 发表于 2013-6-5 01:43
In response to question 2, You should know that you are fitting a ordinary linear model with 4 predi ...

学习了