Back to PLS Help

Bootstrapping question
MAngel
Posted on 04/21/21 22:24:17
Number of posts: 9
MAngel posts:

Dear Randy,

Can I please ask why you think using 100 vs. 1000 iterations in bootstrapping would produce different results? In most PLS papers I've seen, 100 iterations have been used and now the reviewr is asking why not 1000. I've used 1000, some of the LVs are giving me different results for the meancentring. So, trying to answer that question why 100 and not 1000. Any help would be appreciated!

Best,

Maryam

Replies:

Untitled Post
randy
Posted on 04/22/21 07:44:14
Number of posts: 378
randy replies:

quote:

Dear Randy,

Can I please ask why you think using 100 vs. 1000 iterations in bootstrapping would produce different results? In most PLS papers I've seen, 100 iterations have been used and now the reviewr is asking why not 1000. I've used 1000, some of the LVs are giving me different results for the meancentring. So, trying to answer that question why 100 and not 1000. Any help would be appreciated!

Best,

Maryam

Hi Maryam

The number of bootstrap iterations necessary to get stable standard error estimates is a function of sample size. With a smaller sample size (e.g., n=10), 100 iterations is usually enough to hit asymptote in the estimation.  As you get larger N's you need to do more iterations.  I would agree with the reviewer that 1000 would be good and even more if your computer is fast.




Login to reply to this topic.

  • Keep in touch

Enter your email above to receive electronic messages from Baycrest, including invitations to programs and events, newsletters, updates and other communications.
You can unsubscribe at any time.
Please refer to our Privacy Policy or contact us for more details.

  • Follow us on social
  • Facebook
  • Instagram
  • Linkedin
  • Pinterest
  • Twitter
  • YouTube

Contact Us:

3560 Bathurst Street
Toronto, Ontario
Canada M6A 2E1
Phone: (416) 785-2500

Baycrest is an academic health sciences centre fully affiliated with the University of Toronto