log-normal and gaussian distribution - array generation

mresimulator
165 ●32 ●43 ●52

Hi experts!

I wanna generate a random number array of size=N using a log normal distribution. From http://en.wikipedia.org/wiki/Log-normal_distribution i wanna use the parameters mu and sigma.

I know that I must do:

form scipy.stats import lognorm
new_array = lognorm.rvs(......, size=N)

What must I set like parameters (loc, s, scale, etc.) for use mu and sigma distribution parameters.

In the same way: what must I do in

new_array = norm.rvs(......, size=N)

for generate a array of random numbers using a gaussian distribution with parameters mu and sigma?

Waitign for your answers.

Thanks a lot!

add a comment

2 Answers

Sort by » oldest newest most voted

answered 11 years ago

calc314
4191 ●22 ●48 ●113

updated 11 years ago

You might also prefer to use numpy.random. You can get help at:

http://docs.scipy.org/doc/numpy/refer...

For example, try:

import numpy.random as rdm
rdm.normal(4,1,10)
rdm.lognormal(4,2,20)

link

Comments

Thanks a lot!!

mresimulator ( 11 years ago )

add a comment

answered 11 years ago

Shashank

1917 ●31 ●54 ●85

You can convert a normal or log distribution with mu=0 and sigma=1, which is the default to whatever you want. For eg. for normal distribution you can use the following to get a normal distribution with sigma=10 and mu=1

from scipy.stats import norm
R = norm.rvs(size=10)
sigma=10
mu=1
print sigma*R+mu

link

add a comment

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

log-normal and gaussian distribution - array generation

2 Answers

Comments

Your Answer

Question Tools

Stats

Related questions

log-normal and gaussian distribution - array generation savecancel

2 Answers

Comments

Your Answer

Question Tools

Stats

Related questions

log-normal and gaussian distribution - array generation