central limit theorem

[nolsqn]

hi all

i want to apply central limit theorem which states states that the re-averaged sum of a sufficiently large number of identically distributed independent random variables each with finite mean and variance will be approximately normally distributed. any help??

[Coq_Rouge]

Have you tried Google and this search prhase ->central limit theorem labview. I am sure you will find something

[Jon_S.]

nolsqn,

I think I am a little confused as to what you want to do.  Instead of describing the theory could you give an example.  Also check out the Probability and Statistics palette. You can find this in the functions palette.  Select Mathematics»Probability and Statistics.   Take a look at the Statistics express VI in that palette.

[ahmedhamdy2010]

hello,

did you find anything regarding CLT, i also need to implent it on labview and i found nothing

[VItan]

---

@ahmedhamdy2010 wrote:

hello,

did you find anything regarding CLT, i also need to implent it on labview and i found nothing

---

Jon_S.'s comment still applies: What are you trying to do?  How are you trying to applay the Central Limit Theorem?  As far as I know, there is nothing inherentely "implementable" in the CLT.

[ahmedhamdy2010]

thanks for your quick response, i have a sample set of 4400 samples (8 bits integers), i want to extract all the possible combinations of 50 samples from the set and get their average .

which will be 4400C50 possibilities. 

[VItan]

What have you tried?

[ahmedhamdy2010]

---

@VItan wrote:

What have you tried?

---

i have found this vi here in the forums, which gets all the possible combinations of a set of samples. then i could get the average of each row, no problem

the problem is when the sample size increase, as in my case, there is no enough memory exists to hold such a large matrix

[VItan]

Why do you need to do this for every choice?

The nCr VI that you attached has trouble with large values.  It returns the max signed 64 bit integer value of 9223372036854775807 for 4400C50 because the actual number of choices is on the order of 10^117.  There is actually a binomial coefficient DLL call VI in Mathematics -> Elementary & Special Functions -> Discrete Math that you could use to get 4400C50.

With that many choices, making an array out of all of the choices at once will not be feasible due to memory use.

However, even if you had all the memory or even if you just do one at a time to save on memory, it would take an amount of time that doesn't even exist: Assuming you can perform one choose and average in 1 ms - this would take on the order of 10^105 years; the estimated age of the universe is only on the order of 10^10 years...

[ahmedhamdy2010]

---

@VItan wrote:

Why do you need to do this for every choice?

The nCr VI that you attached has trouble with large values.  It returns the max signed 64 bit integer value of 9223372036854775807 for 4400C50 because the actual number of choices is on the order of 10^117.  There is actually a binomial coefficient DLL call VI in Mathematics -> Elementary & Special Functions -> Discrete Math that you could use to get 4400C50.

With that many choices, making an array out of all of the choices at once will not be feasible due to memory use.

However, even if you had all the memory or even if you just do one at a time to save on memory, it would take an amount of time that doesn't even exist: Assuming you can perform one choose and average in 1 ms - this would take on the order of 10^105 years; the estimated age of the universe is only on the order of 10^10 years...

---

yes, you are absolutely right, but i think there must be a way around it. 

yes, it would be ideal to do this for each choice, but i think i can  take only 10^6 choices for good results, i can also try to reduce the samples as much as possible.

with these considerations, is it possible?