Propeller Head Plaza

Technical and scientific discussion of amps, cables and other topics.

Return to Propeller Head Plaza


statistics question

137.78.16.139

Posted on October 8, 2009 at 13:35:39
mike1127
Audiophile

Posts: 27
Joined: June 3, 2009
This isn't directly related to audio, but I know there are some statistics propeller heads here, so thought I would ask. This is part of a simulation task. I need to generate a pseudorandom gaussian distribution of numbers in a programming language that only has a linear distribution of pseudorandom numbers available. Is there a forumla to turn one into the other? This might be something like the integral of the bell curve, not sure.

Thanks,
Mike

Start here, posted on October 9, 2009 at 06:30:38
Tony Lauck
Audiophile

Posts: 3048
Location: Vermont
Joined: November 12, 2007
Contributor
  Since:
February 24, 2009

Tony Lauck

"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar

I have always wondered why we don't use the value of π and enter at the random number computer generated. nt, posted on October 9, 2009 at 16:39:25
Norm
Audiophile

Posts: 14914
Joined: September 6, 2000
a

I don't have an answer for you, but I have always been bemused that no one really has a random numb. generator, posted on October 9, 2009 at 06:25:38
Norm
Audiophile

Posts: 14914
Joined: September 6, 2000
for computers. It draws into question where those in stat. books come from.

No equivalent of GSL library for that language (what is it, BTW)?, posted on October 8, 2009 at 20:38:24
carcass93
Audiophile

Posts: 2634
Location: NJ
Joined: September 20, 2006
Or, alternatively, can C++ DLL be used from that language?

RE: No equivalent of GSL library for that language (what is it, BTW)?, posted on October 9, 2009 at 15:39:02
mike1127
Audiophile

Posts: 27
Joined: June 3, 2009
Actually there is probably a lot of statistical stuff available in modules for this language, which is Haskell, but I was not able to find anything that I understood. I thought it would be a lot easier to implement one equation, assuming it only takes one equation.

Look here - probably something you could use (link)., posted on October 10, 2009 at 07:36:41
carcass93
Audiophile

Posts: 2634
Location: NJ
Joined: September 20, 2006
N/T

RE: Look here - probably something you could use (link)., posted on October 10, 2009 at 17:33:33
mike1127
Audiophile

Posts: 27
Joined: June 3, 2009
Thanks. It's hard to understand how to use that library---I am fairly new to Haskell and it's a pure functional language, which means many things are radically different than other languages. Random numbers are more complex to deal with in pure functional languages, so there are some subtleties I don't grasp.

I guess I'll have to "bite the bullet" and settle down to do a few days of research. Was hoping to knock off this job quickly, but 'tis not to be.

Page processed in 0.025 seconds.