k*******g
发帖数: 13 | 1 To predict y(t), we use two candidate vars x1(t) x2(t) separately and get
two linear regression models:
M1: y(t)=b1*x1(t)+b0
M2: y(t)=b1*x2(t)+b0
The coefficient of determination, R-squared, for M1 and M2 are 0.01 and 0.02
If we run a new regression model with both x1 and x2,
M: y(t)=b1*x1(t)+b2*x2(t)+b0
问：lower bound and upper bound of M's R-squared 是多少？这两个extreme cases
分别会有什么问题？
多谢！ |
J*****n
发帖数: 4859 | 2 如果是考虑random variables而不是sample，那似乎是0.02 - 1。 |
J*****n
发帖数: 4859 | 3 如果是考虑random variables而不是sample，那似乎是0.02 - 1。 |
g*********g
发帖数: 26 | 4 0.02-0.03
0.02说明X2中已经包含了所有X1的信息
0.03说明X1和X2对于解释数据的信息相互不重叠
猜的 哈哈 请指教 |
z****t
发帖数: 78 | 5 It should be [0.02,1]. It is easy to construct data sets for both extreme
cases. R^2 of 0.01 and 0.02 is really low, none of the linear model has any
explanatory power. In the case new R^2 is 1, two samples happen to exactly
fit y(t). But obviously it is overfitted. |
