How to generate random numbers in a Gaussian distribution?

Given A = 1 and B = 5, and knowing that the mean is 2.5 and the standard deviation is 1, I want to generate 1000 random points between A and B using a normal (Gaussian) distribution. I am unsure how to do this in MATLAB efficienctly.

[John D'Errico]

They are not normally distributed if they are bounded as you want. You COULD use a truncated normal distribution however.

help truncate

[Ali Almakhmari]

I see. And what if they are not bounded (assume there is no A and B), how will I do it?

[Voss]

If they are not bounded, you can use randn:

https://www.mathworks.com/help/matlab/ref/randn.html

[Ali Almakhmari]

Is this right? I was trying to follow the second example but they used some function called "chol" which I ignored cause I didn't understand..so I am not really sure.

mean = 2.5;
sigma = 1;
values = repmat(mean,1000,1) + randn(1000,1)*sigma

[Walter Roberson]

values = mean + randn(1000,1) * sigma;

The repmat() is valid, but slower than needed for this purpose.

[Ali Almakhmari]

I see. Thank you all so much!!

Answer 1

[Swaraj]

You can use the randn function in MATLAB.

You can follow the following steps:

1. Generate Random Numbers from a standard normal distribution.
2. Scale and shift them to match the desired Mean and Standard Deviation.

Please go through the code below:

A = 1;
B = 5;
std = 1;
mean = 2.5;

step1Result = mean + std * randn(1000,1);
step2Result = min(max(step1Result,A),B);