I think it was my mistake , I switched the sd and mean position..

On Tue, Nov 12, 2013 at 3:45 PM, Craig DeForest
<[email protected]> wrote:
> I think it was luck that those three were negative.  grandom() gives you zero 
> mean and unit variance, so half the numbers that come out of it are negative.
>
>
> On Nov 12, 2013, at 1:10 PM, mraptor <[email protected]> wrote:
>
>> thanks... my mistake with the byte thing..
>> What I want to do is to generate normally distributed random numbers,
>> where the range is humongous (fe: 2^1000)
>> I just tried :
>>> $n = 1000
>>> p grandom(3) * ((2**$n)/2) + sqrt((2**$n))/2
>> [-1.9906798e+300 -3.9087665e+300 -5.6272303e+300]
>>
>>
>> those seem like a random numbers !! but why negative ?
>> Do you think I should use Math::BigInt
>>
>> On Mon, Nov 11, 2013 at 11:55 AM, Craig DeForest
>> <[email protected]> wrote:
>>> The issue is that grandom doesn't work properly on integer types, since it 
>>> makes a distribution with mean 0 and stddev 1.  It's a bit awkward, but you 
>>> can use
>>>
>>> $vals = (grandom($n) * $stdev + $mean)->rint->byte
>>>
>>> instead.
>>>
>>>
>>> On Nov 11, 2013, at 8:59 AM, mraptor <[email protected]> wrote:
>>>
>>>> Is there are way to generate random values based on Normal distribution.
>>>> The same like rnorm() in R-language.
>>>>
>>>>> round( rnorm(3,mean=2^n/2,sd=sqrt(2^n)/2) )
>>>> [1] 524758 524540 524647
>>>>
>>>> I see there is grandom() funtion, but it requires some sort of lib (ndtri) 
>>>> ?
>>>>
>>>> pdl> p grandom(byte,1)
>>>>
>>>> ndtri domain error
>>>> [0]
>>>>
>>>>
>>>> thanks
>>>>
>>>> _______________________________________________
>>>> Perldl mailing list
>>>> [email protected]
>>>> http://mailman.jach.hawaii.edu/mailman/listinfo/perldl
>>>>
>>>
>>
>> _______________________________________________
>> Perldl mailing list
>> [email protected]
>> http://mailman.jach.hawaii.edu/mailman/listinfo/perldl
>>
>

_______________________________________________
Perldl mailing list
[email protected]
http://mailman.jach.hawaii.edu/mailman/listinfo/perldl

Reply via email to