3 views (last 30 days)

Omer Yasin Birey
on 4 Feb 2019

Edited: Omer Yasin Birey
on 4 Feb 2019

normrnd(0,1)

or

randn(5,1)% 5 is just an example it will create 5*1 matrix adjust the value

Guillaume
on 4 Feb 2019

Comment by Khaled Haj Taieb mistakenly posted as an answer moved here:

thank you for your answer :)
but i get some values more than 1 and less than -1. do the values have to be between (-1,1) ?

Guillaume
on 4 Feb 2019

Reply by Omer Yasin Birey also moved here

Hi Khaled, firstly probably you are trying with low values as parameters such as

randn(5,1)%5 is low here

If you increase this value you probably will see values besides -1 to 1. However, you want mean 0 and std 1, the normal distribution is a kind of distribution, where almost half of the values should be greater than mean and the other almost half should be lower than mean. So with these mean and std values it is not likely to go far away from mean with normally random values. The randomly distributed values will end up positive or negative side of 0 (since it is the mean) and we prefer the distance 1(since it is the std). And the average of total distribution should converge 0. Therefore -1 and 1 is quite likely to happen here. If you want to test normal distribution, you can use other mean and std values such as.

5 + 10*randn(1000,1); % 5 is mean, 10 is std here

In this example most of them will be between -5 to 15. Since the mean is 5 std is 10 here. But of course some nasty boys will go out from this boundries. But we talk about the generalizations here.

Guillaume
on 4 Feb 2019

"but i get some values more than 1 and less than -1"

You need to go back and read the definition of a normal distribution and the definition of standard deviation.

A distribution of mean 0 and standard deviation 1 must have values outside of the range [-1,1], otherwise the standard deviation must be less than 1.

Sign in to comment.

Sign in to answer this question.

Opportunities for recent engineering grads.

Apply Today
## 0 Comments

Sign in to comment.