[R] [FORGED] Testing wether my dataset follows a poisson distribution with R

[Rolf Turner]

On 22/07/20 2:26 pm, Paul Bernal wrote:

> Dear friends,
> 
> I have a sample dataset, which is basically the number of transits through
> a particular waterway, and is on a daily basis.
> 
> MyDat <- dataset$DailyTransits
> 
> What I´d like to do is to test whether MyDat follows a poisson distribution
> or not. What R function could accomplish this?
> 
> Any help and/or guidance will be greatly appreciated,

I presume (your question is a bit vague) that you want to do a goodness 
of fit test of the Poisson distribution to your data.  Doing such a test 
would, I think, involve the assumption that your observations are 
independent and identically distributed (i.i.d.).  Since you appear to 
have a time series of daily counts, this assumption is unlikely to be valid.

Modelling such a time series and investigating whether the underlying 
marginal distributions are Poisson is likely to be a fairly subtle 
problem.  Others may be able to offer more insight here.  However such a 
discussion would be about statistical theory and methodology and  hence 
inappropriate for this list.

If you make the (likely to be untenable) assumption that your data are 
i.i.d. then such a g.o.f. test is probably most appropriately done using 
a chi-squared goodness of fit test.

The function chisq.test() would  help you.  Getting the details right 
may prove a bit tricky, but if you can't handle that, then you should 
probably seek "local" statistical advice.

In fact, you should (in view of the likely lack of independence of your 
data) seek local statistical advice in any case.

cheers,

Rolf Turner

P.S. Please note that it's "whether" not "wether".  (A wether is a 
castrated ram.)  Also "Poisson" should be capitalised.  (It's a bloke's 
name.)

R. T.

-- 
Honorary Research Fellow
Department of Statistics
University of Auckland
Phone: +64-9-373-7599 ext. 88276