Original Sample(O), Sample Mean(M), Standard Deviation(STDEV)

Small talk on SmartPLS that does not correspond to the other forums!
Post Reply
KIHO KWON
PLS Junior User
Posts: 9
Joined: Tue May 15, 2018 1:37 am
Real name and title: Mr. KIHO KWON

Original Sample(O), Sample Mean(M), Standard Deviation(STDEV)

Post by KIHO KWON » Thu May 31, 2018 1:16 pm

Hi.

When I use standard bootstrapping, 'Original Sample(O)', 'Sample Mean(M)' and 'Standard Deviation(STDEV)' are given.

What does 'Original Sample' mean by?

I think 'Sample mean' is the results from all resamples in the bootstrap procedure and 'STDEV' is the results from all resamples in the bootstrap procedure as well. Is it right?

Because I have some errors below, I have to calculate t statistics and p value manually. But I am confused a little bit.

Thanks
KIHO KWON
Attachments
regular bootstrapping result.PNG
regular bootstrapping result.PNG (28.43 KiB) Viewed 281 times

jmbecker
SmartPLS Developer
Posts: 842
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: Original Sample(O), Sample Mean(M), Standard Deviation(STDEV)

Post by jmbecker » Sun Jun 03, 2018 7:34 pm

Original Sample is the value of the parameter in the original sample.

And yes you are right:
Sample Mean is the mean of the estimated parameter over all subsamples. It is only important for bias corrected confidence intervals to calculate the bias.
STDEV is the standard deviation of the estimated parameter over all subsamples. In bootstrapping it equals the standard error and is thus used to calculate the t-statistic (Original/STDEV).
Dr. Jan-Michael Becker, University of Cologne, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Ja ... v=hdr_xprf
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de

KIHO KWON
PLS Junior User
Posts: 9
Joined: Tue May 15, 2018 1:37 am
Real name and title: Mr. KIHO KWON

Re: Original Sample(O), Sample Mean(M), Standard Deviation(STDEV)

Post by KIHO KWON » Tue Jun 05, 2018 2:47 am

Then,
1)How can I calculate P values by using original sample, sample mean, standard deviation, and t statistics?

2) How can I calculate Confidence Interval of 97.5% the values of which are "n/a" as you can see below
Attachments
캡처.PNG
캡처.PNG (19.46 KiB) Viewed 271 times

jmbecker
SmartPLS Developer
Posts: 842
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: Original Sample(O), Sample Mean(M), Standard Deviation(STDEV)

Post by jmbecker » Tue Jun 05, 2018 7:12 am

1) You need to look into the t-table or use a function in another software like excel to compute p-values from t-values. The degrees of freedom is the number of bootstrap samples you have.

2) The easiest way to calculate confidence intervals is the percentile method, where you calculate the 2.5% and 97.5% quantile of the bootstrap parameter distribution that you have in the "Samples" output.
The following excel sheet provides an example how to calculate studentized confidence intervals for mediating models: https://www.pls-sem.net/app/download/15 ... 1485782413
It also demonstrates how to calculate p-values.
Dr. Jan-Michael Becker, University of Cologne, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Ja ... v=hdr_xprf
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de

Post Reply