z**********i
发帖数: 88 | 1 My boss asked me to get the roc curve in r, I never used R before, so all is
new. All I remember is that he mentioned rpart. I know is SAS I need to
perform logistic regression outcome y(normal=0, abnormal=1) on x(a
continuous variable).Then get the roc curve. But have no idea what to do in
R. Anybody can help? Please and thanks!
data is like this:
y x
1 1
0 2
0 3
0 4
1 5
1 8
0 6 |
a********s
发帖数: 188 | 2 I just checked that an R package "Epi" has a function called ROC which can
be used to draw roc curve for logistic regression.
GL! |
d*******o
发帖数: 493 | |
z**********i
发帖数: 88 | 4 library(ROCR), YES. I don't know how to interprete the result.
~~~~~~~~~
Thanks, everybody |
R****n
发帖数: 708 | |
D******n
发帖数: 2836 | 6 还是自己写个AUC的function吧，就几行。
我用那个rocr的居然会报错，不够内存，不知道那些人怎么写的。
【在 R****n 的大作中提到】
: read this ppt. It is quite good!
: http://www.modelia.org/html/081118_ModellingCoursePoland/pdf/11
|
R******d
发帖数: 1436 | 7 一列TPR一列FPR，很容易就算出来了
all is
in
【在 z**********i 的大作中提到】
: My boss asked me to get the roc curve in r, I never used R before, so all is
: new. All I remember is that he mentioned rpart. I know is SAS I need to
: perform logistic regression outcome y(normal=0, abnormal=1) on x(a
: continuous variable).Then get the roc curve. But have no idea what to do in
: R. Anybody can help? Please and thanks!
: data is like this:
: y x
: 1 1
: 0 2
: 0 3
|
h******y
发帖数: 3501 | 8 搭车问一下，什么TEST能检测２个或N个ROC CURVE是否SIGNIFICANT DIFFERENT？
多谢。 |
s*******2
发帖数: 499 | 9 There are at least three packages in R about roc curve area. But I think the
calculation is more accurate in SAS.
is
in
【在 z**********i 的大作中提到】
: My boss asked me to get the roc curve in r, I never used R before, so all is
: new. All I remember is that he mentioned rpart. I know is SAS I need to
: perform logistic regression outcome y(normal=0, abnormal=1) on x(a
: continuous variable).Then get the roc curve. But have no idea what to do in
: R. Anybody can help? Please and thanks!
: data is like this:
: y x
: 1 1
: 0 2
: 0 3
|
q********i
发帖数: 795 | 10 DeLong, E. R., DeLong, D. M., and Clarke-Pearson, D. L. (1988), “Comparing
the Areas under Two or More
Correlated Receiver Operating Characteristic Curves: A Nonparametric
Approach,” Biometrics, 44, 837–845.
【在 h******y 的大作中提到】
: 搭车问一下，什么TEST能检测２个或N个ROC CURVE是否SIGNIFICANT DIFFERENT？
: 多谢。
|
