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 to fit 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
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

Mar 17, 2011 9:55 PM
(2690 views)

Hi All, Could you please advice how to fit both linear regression and polynomial (degree=3) model together? Thank you.

1 ACCEPTED SOLUTION

Accepted Solutions

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

Mar 18, 2011 3:14 AM
(3831 views)

I am not sure exactly what you want to do, but it should be quite straightforward.

In the Fit Model platform select Y variable(s), click "Y". Select linear effect and click "Add", select polynomial effect and enter "3" in the "Degree" field and finally select "Polynomial to Degree" in the "Macros" drop-down menu.

In the Fit Model platform select Y variable(s), click "Y". Select linear effect and click "Add", select polynomial effect and enter "3" in the "Degree" field and finally select "Polynomial to Degree" in the "Macros" drop-down menu.

4 REPLIES

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

Mar 18, 2011 3:14 AM
(3832 views)

In the Fit Model platform select Y variable(s), click "Y". Select linear effect and click "Add", select polynomial effect and enter "3" in the "Degree" field and finally select "Polynomial to Degree" in the "Macros" drop-down menu.

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

Mar 20, 2011 8:44 PM
(2652 views)

MS, Thank you for you advice. I would like to explain more what I want to do.

I have 3 input values (x1,x2 and x3) and the output y variable.

As I plot and fit the relations:

(x1,y) = linear regression,

(x2,y) = linear regression,

(x3,y) = polynomial degree=3.

So I would like to get transfer function for prediction the output Y.

I'm not sure which method should I use to get prediction formula by use fit model or neural net. Could you please advice more?

Thank a lot!

Message was edited by: TT

I have 3 input values (x1,x2 and x3) and the output y variable.

As I plot and fit the relations:

(x1,y) = linear regression,

(x2,y) = linear regression,

(x3,y) = polynomial degree=3.

So I would like to get transfer function for prediction the output Y.

I'm not sure which method should I use to get prediction formula by use fit model or neural net. Could you please advice more?

Thank a lot!

Message was edited by: TT

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

Mar 21, 2011 12:48 AM
(2652 views)

I have only little experience with neural nets, however as you seem to have a clear idea about your model and it's essentially linear I suggest Fit Model platform. Neural Net is more useful if the functional relationships are unknown or highly latent, at least that's my impression.

Anyway, after you set up run your model in Fit Model you can get the prediction formula by choosing Save columns/Prediction Formula under the red triangle in the result window.

Anyway, after you set up run your model in Fit Model you can get the prediction formula by choosing Save columns/Prediction Formula under the red triangle in the result window.

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

Mar 21, 2011 8:17 AM
(2652 views)

Thanks again MS. I will try as your suggestion.