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
- :
- Exporting Fitted Distribution Results to a New Table

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

Oct 13, 2014 12:19 PM
(7102 views)

Hello,

I watched a recent video (Using JMP® to Assess Risk in Financial Predictions by Using Monte Carlo Simulations - YouTube)

At the end of the video a gentleman asked the very question that I had: If you have a LOT of distributions how do I get the results from this window into a table to add rows and thus make a Monte Carlo. I have daily temperature data that I'd like to have a MC done on per day to construct a daily distribution.

I can go about this manually as the video shows - but I know there must be a quicker method.

Any help on this would be most appreciated!

1 ACCEPTED SOLUTION

Accepted Solutions

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

Oct 24, 2014 11:06 AM
(12257 views)
| Posted in reply to message from wzimmerman 10/24/2014 11:03 AM

Hi wzimmerman,

You can save some time, especially for those distributions with many parameters, but selecting "Save Density Formula" under the Red Triangle for the fitted distribution. This isn't exactly what you want, but gets you much of the way there. For example, selecting this next to "Fitted Normal" saves the following to a formula column:

Normal Density( :height, 62.55, 4.24233849397192 )

This is very close to what you want, so you could go into the formula and replace "Normal Density" with "Random Normal" and delete the reference to the original column, leaving:

Random Normal( 62.55, 4.24233849397192 )

This will work for more complicated distributions. For example, with a Johnson Su, "Save Density Formula" will save:

Johnson Su Density(

:height,

67.8038036468075,

3.96840876806022,

78.5033678146444,

0.0000011762144700618

)

which needs the same change to be the formula you want:

Random Johnson Su(

67.8038036468075,

3.96840876806022,

78.5033678146444,

0.0000011762144700618

)

I hope this helps!

Julian

8 REPLIES

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

Bump.

Does no one have a solution to this?

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

You can right click on any report table and choose "Make into Data Table" or "Make Combined Data Table" to get the results into a data table.

I think I need some more details about exactly how you want to do your simulation to help you with how to take that information into the simulation.

-Jeff

-Jeff

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

Oct 24, 2014 8:03 AM
(6822 views)
| Posted in reply to message from Jeff_Perkinson 10/15/2014 05:03 PM

Hi Jeff,

I'm looking to nab the distribution formulas. For example if I open a table and make a formula in a column: Random Normal (x, y) and then add rows I'll end up with a Monte Carlo simulation of sorts as it will pick a random number along the distribution I just specified.

If I could do that or something similar by taking the distribution formulas or results to another table to create a distribution for each day of data.

The video shows the way to do this with few variables and at the end of the video someone from the audience asks - well how do you do this with many variables. The class decides to do this after the lecture and the video ends without answering it...

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

Oct 24, 2014 11:06 AM
(12258 views)
| Posted in reply to message from wzimmerman 10/24/2014 11:03 AM

Hi wzimmerman,

You can save some time, especially for those distributions with many parameters, but selecting "Save Density Formula" under the Red Triangle for the fitted distribution. This isn't exactly what you want, but gets you much of the way there. For example, selecting this next to "Fitted Normal" saves the following to a formula column:

Normal Density( :height, 62.55, 4.24233849397192 )

This is very close to what you want, so you could go into the formula and replace "Normal Density" with "Random Normal" and delete the reference to the original column, leaving:

Random Normal( 62.55, 4.24233849397192 )

This will work for more complicated distributions. For example, with a Johnson Su, "Save Density Formula" will save:

Johnson Su Density(

:height,

67.8038036468075,

3.96840876806022,

78.5033678146444,

0.0000011762144700618

)

which needs the same change to be the formula you want:

Random Johnson Su(

67.8038036468075,

3.96840876806022,

78.5033678146444,

0.0000011762144700618

)

I hope this helps!

Julian

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

Also, if you have several continuous fits (and I'm guessing you will) you can hold down the control key (or command if you're on the mac) before selecting "Save Density Formula" and this will save, at once, all the density formulae in that platform output (this "broadcasts" the command). I suspect you are familiar with this technique, but if not I wanted to make sure you knew since it's a great time saver.

Julian

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

It does get me most of the way there - but going through 300+ distributions to change a word is not intuitive. How do we suggest to SAS/JMP to get an additional option to make this change for us? It would seem like an easy addition, and would make non-programmers' lives easier.

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

With that many columns it would certainly be a pain to change them all! Through some scripting we might be able to eliminate the bulk of the work, but I agree that it would be helpful to have something in the software already. I'll be happy to submit that feature request on your behalf.

Julian

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

Thanks!