Now I'm interested test statistics for Two sample in high dimensional data.
I need real data set be 2 sample (eg. Cancer and Normal, Dead within 2 years and after 2 years) and each sample is high dimensional data (n1,n2 > p).
next one, each sample has sample size of this data set should be exceed 100 (n1,n2 > 100).
Please recommend me.
Thanks in advance for any help that you are able to provide.
Paranut Sukcharoen,
Your question lacks so much detail (e.g., the objective of the analysis) that it's extremely difficult, if not impossible to offer advice. And, I typically see high dimensionality defined as N ≤ p, where N is sample size and p is number of parameters. I've also heard the "rule of thumb" for high dimensionality is p ≥ 3-4, but I think this is more in regards to classic statistical methods and their interpretation.
I suggest taking more time to think about what you hope to accomplish (i.e., describe, compare, predict, etc.) with your data and analysis, then sharing this with the community. If it's just toy data you are after, have a closer look around the web there's likely a toy data set out there somewhere.
Good luck,
Caleb
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size