statistics - What R packages are available for binary data that is both correlated and clustered? -


I am working on a project that I have ever done before. I have two tests with binary results which will be given for the same sample, which is drawn from a clustered population (i.e., some subjects will be from the same family). I want to compare the ratio of positive test results, but clustering makes the test of McNean inappropriate, so I'm studying in alternative ways. Claustening Adjusted McNimer Options by GE, Rao and Scott (1992), Elyessee and Donner (1991), and Obukhovsky (1998) and 2) GE

Do you know about any implementation of Rao-Obukhovsky genealogy in R (or, I think, SAS)? Finding GEE is easy, but have you experienced a positive or negative with a particular package?

Thank you in advance for your help - tell me whether any explanation is necessary or not.

You can always use a clustered bootstrap, the responses to families that you believe are independent That is, keep the family together, when you remember again, compute p2 - p1 for each sample, calculate the amount of 2.5% above and below, after 1000 iterations. This will give you bootstrap to 95% confidence interval. Alternatively, calculate the fraction of the sample greater than zero or whatever your concept is, till the number of families is not small, there are very good qualities in the process.

Instead of trusting any package, it is easy to do this with R in hand.


Comments

Popular posts from this blog

sql - dynamically varied number of conditions in the 'where' statement using LINQ -

asp.net mvc - Dynamically Generated Ajax.BeginForm -

Debug on symbian -