I have read a paper described an analysis of using support vector regression. In the paper it mentioned:

It is worth mentioning that, in our implementation, we subtract the mean value of the features from the corresponding target variable before perform the training for the support vector regression. In prediction, this mean value will be added to the predicted output target variable value.

It seems to me that they are trying to do a centering of data before applying regression. My questions are:

  1. Why the centering process is necessary for reasonable regression?
  2. Should we also subtract the mean value of the features to each feature value as well? (In the paper, they seems only subtract the mean from the target variable.)
  3. Do we need to add the mean value back to the predicted output?
  4. There is another procedure called 'Standardization'. In which cases we should do centering and when should we use standardization?

In general, just try to figure out how to do the pre-processing of the data correctly and properly before put them into regression. Thanks. A.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.