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
- :
- How can I get MLE for a custom model?

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
- Permalink
- Email to a Friend
- Report Inappropriate Content

May 28, 2016 11:55 PM
(1087 views)

I'm new to JMP so sorry if this is simple. I've read the user manual over and over and cant find the info I need.

I want to find parameters for a custom distribution ideally, or at least use the exponential distribution. Then I want to use it to get MLE for some custom models. I found that I can get MLE with exponential distribution using the "fit model" option and "generalized regression", but I have NO idea how to get it to fit my custom model. I figured out how to fit a simple model, like log(variable), but I don't know how to add parameters to the model.

Alternatively, I can fit my custom model using "nonlinear model" fitting option, but not using MLE on my chosen distribution.

Help please?

1 REPLY

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

Jun 2, 2016 11:12 AM
(994 views)

Hello,

Thanks for using JMP.

I think you should be able to use the Nonlinear platform to fit a custom model. Here's an example of using the Nonlinear platform to fit a logistic regression model: Logistic Regression Example in Nonlinear

The Nonlinear platform fits models by maximizing the likelihood function, so it is providing an MLE solution.

Hope that helps,

Michael

Michael Crotty

Sr Statistical Writer

JMP Development

Sr Statistical Writer

JMP Development