I am trying to consolidate variables in different units. Could someone kindly tell me what could be the best normalization technique?
This is a rather general question. There are many normaization (or transformation) tehcnique available. The most common one is to transform your variables into a unifying scale, basically by normalizing each variable by their measure of dispersion (e.g. standard deviation) based on your assumption or knowledge about the distribution of the data. Another way is to transform each variable into a unifying 0-1 scale (e.g by using their own range). Of course there is always trade-off, because you will always lose some degree of the actual data variation.
This is a rather general question. There are many normaization (or transformation) tehcnique available. The most common one is to transform your variables into a unifying scale, basically by normalizing each variable by their measure of dispersion (e.g. standard deviation) based on your assumption or knowledge about the distribution of the data. Another way is to transform each variable into a unifying 0-1 scale (e.g by using their own range). Of course there is always trade-off, because you will always lose some degree of the actual data variation.
It would help us if you told us what you are still trying to normalize, and why.
What are your units?
See
http://en.wikipedia.org/wiki/Normalization
http://en.wikipedia.org/wiki/Normalization_%28statistics%29
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics