In this paper, we investigate a non-parametric approach to compare two groups in microarray data.
This is done using a threshold penalized-distance likelihood function, which is made up of a penalty and a suitable threshold distance, and is applicable...
In this paper, we investigate a non-parametric approach to compare two groups in microarray data.
This is done using a threshold penalized-distance likelihood function, which is made up of a penalty and a suitable threshold distance, and is applicable when sample size is small or when the data is not normally distributed.
We also use this function to classify new data. This is based on objects that are identified as differences between the two groups, not for all objects. We also study a real data application to illustrate our methods.