g*******a
发帖数: 443 | 1 Graham regresses variable Y on four independent variables: X1, X2, X3 and
X4. He obtains the following output:
图片1
Graham also tries a regression of Y against X1 and obtains the following
results:
图片2
Answer the following questions:
a) Why do you think there is not a big difference between the R-Square of
the two regressions?
b) Graham plots the variable X1 against the variable X2 and obtains the
following graph. What do you think this means? How does it affect the
regression?
图片3
c) Would you conclude that there is no significant relationship between Y
and X2?
d) Which of the two regressions would you use to predict Y? | l******r
发帖数: 18699 | 2 从图三可见基本上X1＝X2
估计X3和X4跟X1有很大的multi-collinearity
所以Y~X1+X2+X3+X4跟Y~X1的model fit结果差不多，故R^2差不多
当然full model不是个好model因为里面包含太多相关变量，
参数估计的准确度是个问题
这从图一X2,X3,X4的p-value比较大可以说明它们不是重要significant变量
of
【在 g*******a 的大作中提到】
: Graham regresses variable Y on four independent variables: X1, X2, X3 and
: X4. He obtains the following output:
: 图片1
: Graham also tries a regression of Y against X1 and obtains the following
: results:
: 图片2
: Answer the following questions:
: a) Why do you think there is not a big difference between the R-Square of
: the two regressions?
: b) Graham plots the variable X1 against the variable X2 and obtains the
| g*****p
发帖数: 3403 | |
|
