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
- :
- Bootstrapping Question

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

Aug 22, 2014 10:18 AM
(2763 views)

When using the Distribution Platform to use bootstrapping, what is the default size of the individual bootstrap samples? Is it possible to select the size of the samples?

Steve

1 ACCEPTED SOLUTION

Accepted Solutions

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

Aug 22, 2014 3:09 PM
(4803 views)

I think the question was about the "sample size", not the number of samples.

My understanding of the bootstrap method is that the sample size would be the number of data points that you have. So in Lou's example, the sample size would be 40. That sounds kind of weird when you think you are taking a "sample". How can you take say 100 samples if each sample contains all 40 data points? The answer is that each sample differs because the sampling is being done **with replacement**. Therefore in any one sample, some data points may appear multiple times and others may not appear at all.

Dave

-Dave

3 REPLIES

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

Aug 22, 2014 1:58 PM
(2623 views)

The bootstrapping capability is a JMP Pro capability. The default number of samples is 100 but one can set it to the desired number through the use of the dialog box that appears when it is enabled.

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

Aug 22, 2014 3:09 PM
(4804 views)

I think the question was about the "sample size", not the number of samples.

My understanding of the bootstrap method is that the sample size would be the number of data points that you have. So in Lou's example, the sample size would be 40. That sounds kind of weird when you think you are taking a "sample". How can you take say 100 samples if each sample contains all 40 data points? The answer is that each sample differs because the sampling is being done **with replacement**. Therefore in any one sample, some data points may appear multiple times and others may not appear at all.

Dave

-Dave

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

Aug 23, 2014 5:11 AM
(2623 views)

Thank-you Dave.

Steve