Turn on suggestions

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

Showing results for

- JMP User Community
- :
- Discussions
- :
- Discussions
- :
- Beta Binomial Generalized Regress Variables

Topic Options

- Start Article
- 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
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Beta Binomial Generalized Regress Variables

Feb 21, 2018 3:07 PM
(631 views)

I have a dataset that is constructed like this:

ID | Replicate | Y | X1 | X2 |

A | 1 | 0 | .78 | .33 |

A | 2 | 0.25 | .44 | .09 |

B | 1 | 0.25 | .76 | .66 |

B | 2 | 1 | .31 | .39 |

C | 1 | 0 | .11 | .48 |

C | 2 | 0 | .07 | .22 |

The beta binomail distribution seems appropriate because my response variable (Y) is porportional. Thus, I am interested in looking at the relationship of the two X variables to Y. In order to use a beta binomial regression in jmp, you need to:

The beta binomial distribution requires a sample size greater than one for each observation. Thus, the user must specify a sample size column. To insert a sample size column, specify two continuous columns as Y in this order: the count of the number of successes, and the count of the number of trials. The link function for p is the logit.

Has anyone ever used this type of analysis and understands how to arrange the dataset to perform it correctly? Thank you for your help!

3 REPLIES 3

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Re: Beta Binomial Generalized Regress Variables

Yes, JMP requires two columns in the Y role. The order of entry in the Y role is important. The first data column is a count of successes (as you define success) and the second column is the count of the total observed (successes plus failures). So, for example, you might have observed 10 cases but none of them were successes when ID=A and Replicate=1. Instead of a proportion of 0, enter a count of 0 and a size of 10. You need to replace the proportional Y in your data table with two data columns.

Learn it once, use it forever!

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Re: Beta Binomial Generalized Regress Variables

Thank you for your help! That is much more clear!

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Re: Beta Binomial Generalized Regress Variables