s******e
发帖数: 1073 | 1 variable x1, x2去predict outcome Y
得出来的model中, 应该怎样分析参数呢?
能说x1 contribute a% percent of outcome and X2 contribute b% of outcome? |
w*******9
发帖数: 1433 | 2 Interpreted as the log odds ratio. |
k*****u
发帖数: 1688 | 3 x2不变得时候，x1增加一个unit，相应的事件的odd ratio增加 exp(x1的系数)
ucla ats上面有很详细很好的教程 |
j******4
发帖数: 6090 | 4 你可以单独fit一下X1，然后看adjusted R square是多少，比如是0.4，那就是说Y里面
40％的variation can be explained by X1。
同理你再单独fit一下X2就可以了。 |
a*z
发帖数: 294 | |
w*******9
发帖数: 1433 | 6 R square for generalized lm does not have that interpretation, even after
adjustment.
【在 j******4 的大作中提到】
: 你可以单独fit一下X1，然后看adjusted R square是多少，比如是0.4，那就是说Y里面
: 40％的variation can be explained by X1。
: 同理你再单独fit一下X2就可以了。
|
C******y
发帖数: 2007 | 7 logistic得用pseudo r2
【在 j******4 的大作中提到】
: 你可以单独fit一下X1，然后看adjusted R square是多少，比如是0.4，那就是说Y里面
: 40％的variation can be explained by X1。
: 同理你再单独fit一下X2就可以了。
|
j******4
发帖数: 6090 | 8 May I ask for a correct interpretation of the R-squared value from GLM?
If I didn't make myself clear, maybe the following quote can help:
"It(R squared value) is the proportion of variability in a data set that is
accounted for by the statistical model."
Wiki page:
http://en.wikipedia.org/wiki/Coefficient_of_determination#cite_
Ref:
Steel, R. G. D. and Torrie, J. H., Principles and Procedures of Statistics,
New York: McGraw-Hill, 1960, pp. 187, 287.
【在 w*******9 的大作中提到】
: R square for generalized lm does not have that interpretation, even after
: adjustment.
|
p*******e
发帖数: 746 | |
w*******9
发帖数: 1433 | 10 Sorry I wish I could help you and I will be very happy if someone could
teach both of us on this point.
My understanding is that R^2 and numerous adjusted/generalized R^2s are all
designed with the desire to assess the goodness of fit by a unitless number
between 0 and 1. They are generally not associated with the proportion of
variation. Only in OLS (with unknown intercept) the R^2 is interpreted as
the proportion of variation explained off by the covariates.
Since there are more popular goodness of fit measures for logistic
regression, I tend to avoid generalized R^2. If you really want to say
something about the predictive ability of X1 and X2, it might be useful to
report the adjusted R^2 or AUCs (area under curves) for the logistic
regressions on x1 and
x2 respectively.
is
,
【在 j******4 的大作中提到】
: May I ask for a correct interpretation of the R-squared value from GLM?
: If I didn't make myself clear, maybe the following quote can help:
: "It(R squared value) is the proportion of variability in a data set that is
: accounted for by the statistical model."
: Wiki page:
: http://en.wikipedia.org/wiki/Coefficient_of_determination#cite_
: Ref:
: Steel, R. G. D. and Torrie, J. H., Principles and Procedures of Statistics,
: New York: McGraw-Hill, 1960, pp. 187, 287.
|
