- #1

- 6

- 0

- Thread starter Weirdzzl
- Start date

- #1

- 6

- 0

- #2

chiro

Science Advisor

- 4,790

- 132

Usually normalizing distributions involves calculating (X-mu)/sigma where if X is Normally distributed, then the result will give N(0,1) or a normalized normal.

Since your data is all negative it is not likely (especially if the sample size is big) that your distribution is normal.

If you want to make it normal you need to transform your data by some function y = f(x) where you apply the function f to all your sample points.

Doing this will depend on the nature of the distribution you have for your sample and what kind of distribution (i.e. population model) you assume the sample belongs to).

This requires graphical attributes of the distribution as well as knowledge about statistics, the nature of the data, and the context of the system that the data belongs in.

In short, it is not a trivial matter.

- Last Post

- Replies
- 1

- Views
- 2K

- Last Post

- Replies
- 4

- Views
- 4K

- Last Post

- Replies
- 13

- Views
- 4K

- Last Post

- Replies
- 4

- Views
- 4K

- Replies
- 1

- Views
- 2K

- Last Post

- Replies
- 2

- Views
- 476

- Replies
- 1

- Views
- 2K

- Replies
- 7

- Views
- 251K

- Replies
- 6

- Views
- 1K

- Replies
- 5

- Views
- 46K