Generate a list of numbers that follow a Poisson distribution?

Hi everyone,

Does anyone know how this can be done or or have any tips? Would it be necessary to work with the Python Editor?

Replies to This Discussion

Permalink Reply by peter fotiadis on April 19, 2016 at 7:02am

Like this?

Attachments:

Poisson_PMF_CDF_V1.gh, 126 KB

Permalink Reply by John Cyrillos on April 20, 2016 at 1:08pm

Looks good! Thanks a lot!

Permalink Reply by peter fotiadis on April 20, 2016 at 2:29pm

And the trad thing (the update,that is)

Attachments:

Poisson_PMF_CDF_V1A.gh, 128 KB

Permalink Reply by John Cyrillos on April 22, 2016 at 3:39am

There seems to be a problem with the file you sent. If "count" is set larger than 32, there is an error. I changed all the decimal types to double, but then a lot of zeros and +infinity appear in the results.

Permalink Reply by peter fotiadis on April 22, 2016 at 3:44am

OK, I'll investigate and do the usual thing: the V2

Permalink Reply by peter fotiadis on April 23, 2016 at 4:53am

Maybe this can shed some light on that matter.

PS: count and lambda are "related" somehow > see bottom-most C# for what actually mean.

Attachments:

Poisson_PMF_CDF_V2.gh, 125 KB

Permalink Reply by peter fotiadis on April 23, 2016 at 5:34am

Added value ceiling options that rather make some sense more.

Attachments:

Poisson_PMF_CDF_V2A.gh, 125 KB

Permalink Reply by John Cyrillos on April 25, 2016 at 8:06am

If you check the outputs PMFlist and CDFlist on your file, at around 15 and further down, you'll see that the numbers are very wrong. Having lambda at 2.82 and getting PMF multiple times at 18.0 and CFD at 30.8.

Permalink Reply by John Cyrillos on April 27, 2016 at 7:23am

Also, in the PMFlist, some unexpected zeros appear where they shouldn't, but I guess that's due to the limitations of the bits used by the double variable type and can't be fixed. Any help, however, will be appreciated, regarding the previous problem.

Permalink Reply by peter fotiadis on April 27, 2016 at 8:47am

Well ...I've noticed these freaky results already and that was the reason for adding the tolerance "filter" and the ceiling/remap options.

However as you can easily testify the equations are rather simple, by the book and nothing is "odd".

Have no idea why the values go bananas (and/or "oscillate") > Plan Z > maybe ask Poisson ? (spirit rapper required > what had the great man in mind ? you tell me).