turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- Analytics
- /
- Stat Procs
- /
- How to calculate P value for Poisson distribution?

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

06-08-2016 10:13 PM

Can anybody please let me know how to calculate P value for a Poisson distribution? Thanks.

Accepted Solutions

Solution

06-09-2016
09:49 PM

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

06-09-2016 09:47 PM

All Replies

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

06-08-2016 10:30 PM

I believe that is using the CDF, cumulative distribution function.

Solution

06-09-2016
09:49 PM

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

06-09-2016 09:47 PM

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

06-10-2016 06:19 AM

You accidentally marked your own replay as the solution.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

06-09-2016 06:41 AM

If you have an observed value of some test statistics that follows a Poisson distribution, then Reeza is correct and you can use the CDF function:

lambda=10; obsValue=15;

OneSidePVal = 1 - cdf("Poisson", ObsValue, lambda);

However, since most statistics do not follow a Poisson distribution, maybe you are asking about fitting a Poisson distribution to data and obtaining a p-value for the goodness-of-fit test? In that case, you should use PROC GENMOD to fit the distribution as shown in the article "Fitting a Poisson distribution to data in SAS". (You can also use COUNTREG.)

Unfortunately, the statistics that come out of PROC GENMOD do not include p-values, but you can use PROC FREQ to compute a chi-square statistics that compares the observed and expected values in each category. The details are shown in the article "Testing the fit of a discrete distribution." If you are interested, there are also graphical techniques for evaluating the fit.

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

06-09-2016 09:51 PM

Thank you so much. Really appreciate. Best regards.