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
- :
- Is it possible to automate building of 1000+ linea...

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

May 6, 2013 7:50 AM
(1547 views)

Hi everyone.

I have a daunting task of finding out the best a linear mixed mixed model using 350+ variables. I know you could do a fit model with stepwise mode but JMP does not allow this function with random effect component. So we are currently building forward selecting the independent variables to build all possible combination of models semi-automatically using a loop function in the script. The downside is I have to manually sift through the reports and copy+paste the "Summary of Fit" and "Parameter Estimates" with the VIF information into an excel file.

My question is **does JMP has script capability to selectively save the results of the report based on a condition that all p-values of each variable is less than 0.05?** The overall construct would be based on 4 independent variables A, B, C and D and 1 dependent variables, Z would be random effect, E would be something like this;

Test E = A + Z(random)

if all term p-value(s) is/are less than 0.05 --> save "Summary of Fit" and "Parameter Estimates" with the VIF column to Results File (excel or word file)

E = A + B+ Z(random)

if all term p-value(s) is/are less than 0.05 --> save "Summary of Fit" and "Parameter Estimates" with the VIF column to Results File (excel or word file) |

E = A + C+ Z(random)

if all term p-value(s) is/are less than 0.05 --> save "Summary of Fit" and "Parameter Estimates" with the VIF column to Results File (excel or word file) |

E = A + D+ Z(random)

E = B+ Z(random)

E = B + C+ Z(random)

E = B + D+ Z(random)

E = C+ Z(random)

E = C + D+ Z(random)

E = D+ Z(random)

end

Any lead on this problem would save me a lot of time and grief. Thank you in advance.

Woong

1 ACCEPTED SOLUTION

Accepted Solutions

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

Solution

This should be very doable. Here are some tips for getting started. Check out the Platform Example starting on p. 339 of the JMP 10 scripting guide. This goes through how to pull individual values out of a table in a report. Once you identify a hit you can use the "<<Make Into Data Table" command to capture the entire table.

2 REPLIES

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

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

May 6, 2013 2:09 PM
(1179 views)