h******e
发帖数: 1791 | |
p******d
发帖数: 18 | 2 principle component
cluster, etc |
Y***Y
发帖数: 180 | 3 PCA
【在 h******e 的大作中提到】
: 谢谢。
|
w**********y
发帖数: 1691 | 4 Always analyze data by graph, table or other statistical summary before
any modelling. Choose some potential available models based on the
experts’ experience and analysis.
Then Modelling by:
1. Ridge regression-biased estimator of beta (degree of biases is
controlled by a constant choosen by urself, and VIF(variance influence
factor) could helps you to choose this constant)
2. PCA (Principal component analysis) – create new variables that is
independent. But problem is how to explain the |
w*****e
发帖数: 806 | 5 3x!! Really good post..
【在 w**********y 的大作中提到】
: Always analyze data by graph, table or other statistical summary before
: any modelling. Choose some potential available models based on the
: experts’ experience and analysis.
: Then Modelling by:
: 1. Ridge regression-biased estimator of beta (degree of biases is
: controlled by a constant choosen by urself, and VIF(variance influence
: factor) could helps you to choose this constant)
: 2. PCA (Principal component analysis) – create new variables that is
: independent. But problem is how to explain the
|
t********y
发帖数: 469 | 6 one can check for multicollinearity by means of the correlation matrix. In
such a matrix when the correlation coefficient between two explanatory
variables is above 0.8 one needs to be aware of possible collinearity. If
the correlation coefficient is above 0.95 the problem is really serious.
A diagnostic approach to check for multicollinearity after performing
regression analysis is to display the Variance Inflation factor (VIF).
VIF is a measure of how much the variance of an estimated regressi |
t********y
发帖数: 469 | 7 Checking for Multicollinearity Using SAS
www-personal.umich.edu/~kwelch/finan/day3_finan_collin.doc |
w*****e
发帖数: 806 | 8 3x a lot. 这个prof好多sas在各个方面的应用啊。
看过doc文件之后觉得讲的很有条理。
【在 t********y 的大作中提到】
: Checking for Multicollinearity Using SAS
: www-personal.umich.edu/~kwelch/finan/day3_finan_collin.doc
|
s******1
发帖数: 26 | 9 其实PCA 也不一定好，需要对新的变量有比较合理的解释。Ridge也要看。学回归的时
候好像也要看是否是 model based collinearity or variable based collinearity.
还有可以作中心差。总之 google也可以goo很多出来的。好运了。 |
