# How to normalize a probability distribution

Printable View

• Nov 2nd 2012, 06:27 PM
tangibleLime
How to normalize a probability distribution
This is just a straightforward question that I can't find an answer to anywhere on the internet. I keep seeing examples of "normalizing a vector" but I don't think that's what I want.

I have this probability distribution:
{1/8, 1/16, 1/32}

And I need to normalize it, which I guess means all of the members of the distribution sum to one. I don't know why, but it's apparently important. This isn't a homework question or anything, but I do need to know how to normalize.

From just moving numbers around, I can see that adding them all together then dividing each member by the total sum appears to do the trick. Is this correct? Is this what "normalizing" is?

{1/8, 1/16, 1/32}
--> {(1/8)/(1/8+1/16+1/32),(1/16)/(1/8+1/16+1/32),(1/32)/(1/8+1/16+1/32)}
--> {4/7, 2/7, 1/7}

4/7 + 2/7 + 1/7 = 1.

Any help is appreciated, thanks.
• Nov 2nd 2012, 09:22 PM
Salahuddin559
Re: How to normalize a probability distribution
Yes, that is correct. That is the general method of normalizing a probability distribution. You add all (or integrate all over the relevant interval) and then divide the distribution function by that sum (integration result).

Salahuddin
Maths online