I'm new to data analysis and data mining. Often in the papers I'm reading, they use the term "high dimensional multivariate data set." Currently, my task is to detect an outlier and visualize the same from a large complex data set. But how does one find out whether I have a multivariate high dimensional data set or not?

  • $\begingroup$ Apropos your real concern, you might find some useful information by searching our site for multivariate outlier. $\endgroup$
    – whuber
    Nov 20 '13 at 14:33
  • $\begingroup$ In high-dimensional space, due to the "curse of dimension", every observation can be considered an outlier. So you need to think hard about the purpose of such analysis... $\endgroup$
    – Michael M
    Nov 20 '13 at 15:31

A high dimensional multivariate data set would simply be a data set with lots of variables. These days, most data sets qualify. Exactly how many variables makes it "high" is not, as far as I know, generally agreed to.

  • 4
    $\begingroup$ "High" presumably means no more than "challenging" to one or more of the hardware, software, programmer or user. I agree with Peter's implication that there is not an agreed technical definition. $\endgroup$
    – Nick Cox
    Nov 20 '13 at 14:30

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.