[R] ROC plot with SEC and SPC as input

Hi, I am working on a ROC plot problem. Thanks in advance for any help. We run 20 cross validation (CV) on our data for each customized threshold, i.e. from 0.1 to 0.9. Then for these 9 threshold values, we get the overall performance, that is we are able to get overall sensitivity and specifi

[R] how to customize row order from clustering in a heatmap?

Hi, I am working on a heatmap and clustering all the rows with the method of 'ward'. But I notice that my rows are in a weird order which is not what I want / expect. Could anyone help on this? My questions are with the figure below. Thanks in advance http://r.789695.n4.nabble.com/file/n360381

[R] experience with nvmax(max size of subsets to examine)

ith 2-variables by setting nvmax=1. Similarly, when setting nvmax=2 and using more than 1000 variables, I also got models with 3-variables. It could be my understanding on nvmax is wrong or some other reason. Does anyone have some idea on this? Thank you. CZ -- View this message in context:

Re: [R] Does R have function/package works similar to SAS's 'PROC REG'?

Hi, Josh, What we are doing is, we have a microarray data set with 2000 genes and roughly 60 samples split 2:1 cancer:normal. So we essentially have one binary response and 2000 continuous predictors. We want to use this to develop an ensemble-based classifier method in which the members of the

[R] Does R have function/package works similar to SAS's 'PROC REG'?

e method is accurate for data that are reasonably scaled and not too collinear. The sweep algorithm is also used in many places in the model-selection methods and the RSQUARE method uses the leaps-and-bounds algorithm by Furnival and Wilson (1974). Thanks. Thanks. CZ -- View this message i

[R] subsets with a small cardinality for variable selection

Hello, I am working on a variable selection problem and would like to have some suggestions. Thank you. In my data, the number of observations/samples is much less than the number of variables. And I am not interested in generating only a few models, instead I will need a couple of hundred m

[R] nvmax in REGSUBSETS function

of 2; and when my nvmax=1, it regresses on 2, which is weird to me. I wonder whether my setting was incorrect or my understanding on nvmax was wrong. Thank you. CZ -- View this message in context: http://r.789695.n4.nabble.com/nvmax-in-REGSUBSETS-function-tp2956926p2956926.html Sent from the

Re: [R] input matrix for leaps algorithm

The regsubsets does work. I also notice the bestglm, will see how it work too. Thank you. -- View this message in context: http://r.789695.n4.nabble.com/input-matrix-for-leaps-algorithm-tp2954453p2956866.html Sent from the R help mailing list archive at Nabble.com.

[R] input matrix for leaps algorithm

Hello, I was trying to use the leaps algorithm for my variable selection. I tried different matrices calculated from multiple linear regression, linear discriminant analysis, and the basic correlation matrix as inputs to the leaps algorithm. But I always got the error message - the matrix is no

Re: [R] Draw a perpendicular line?

Thank you all for your suggestions and especially the codes. Now I am able to solve it by finding the intercepts and slopes for both lines first, and then I can find the coordinates of x and y at the intersection. At first I forgot how to use C to get the intercept of line CD... Thanks again.

[R] Checking a point behind two polygons

Hi, I am working on a problem of checking whether a point is behind two convex hulls. By 'behind', I mean a point which is outside of two convex hulls can't reach either of these two hulls without reaching the other. My idea is to find the segments between the point and the vertices of a hull

[R] Draw a perpendicular line?

Hi, I am trying to draw a perpendicular line from a point to two points. Mathematically I know how to do it, but to program it, I encounter some problem and hope can get help. Thanks. I have points, A, B and C. I calculate the slope and intercept for line drawn between A and B. I am trying to