Bootstrapping (5000 sub-sample)

Research topics can be discussed in this area.
Post Reply
Choukri
PLS Expert User
Posts: 114
Joined: Thu Apr 16, 2015 2:11 pm
Real name and title: Choukri MENIDJEL

Bootstrapping (5000 sub-sample)

Post by Choukri » Tue Apr 21, 2015 7:17 am

Hi everyone,

I have found that the researchers have used the following expretion in their articles, [based on t(4999), one-tailed test], why do they not use this one [based on t(5000), one-tailed test]

I look forward to hear your responds
Thank you

User avatar
Hengkov
PLS Super-Expert
Posts: 1619
Joined: Sun Apr 24, 2011 10:13 am
Real name and title: Hengky Latan
Location: AMQ, Indonesia
Contact:

Re: Bootstrapping (5000 sub-sample)

Post by Hengkov » Tue Apr 21, 2015 3:34 pm

Hi,

I do not understand your question. If it is a T-statistic, then the value of 4999 will be compared with the value of one-tailed t-table. So, not a one-tailed value 4999 (this value is never as big as that for the one-tailed significance level and DF whatever).

Note: Value T-statistic is not the same as the number of resampling.

Greetings,

Choukri
PLS Expert User
Posts: 114
Joined: Thu Apr 16, 2015 2:11 pm
Real name and title: Choukri MENIDJEL

Re: Bootstrapping (5000 sub-sample)

Post by Choukri » Tue Apr 21, 2015 4:49 pm

Hi, thank you Professor for your help

I took the following expretion from an article:
[ns: not significant (based on t(4999), one-tailed test) t(0.05, 4999) = 1.645; t(0.01, 4999) = 2.327; t(0.001, 4999) = 3.092].

jmbecker
SmartPLS Developer
Posts: 820
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: Bootstrapping (5000 sub-sample)

Post by jmbecker » Wed Apr 22, 2015 9:15 am

The authors report the degrees of freedom used for the t-test. The degrees of freedom for a t-test based on bootstrapped SE is the number of resamples -1. Hence, the authors have apparently used 5000 resamples.
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

Choukri
PLS Expert User
Posts: 114
Joined: Thu Apr 16, 2015 2:11 pm
Real name and title: Choukri MENIDJEL

Re: HTMT criterion

Post by Choukri » Wed Apr 22, 2015 2:08 pm

Hi,
Thank you for th respond, just one more question

The following matrix is a part of the full matrix (of my research) for assening desciminant validity by HTMT, I utilized the same forme of Matrix that Henseler et al. (2015) have used in their articles.

How should I interprete it?
Is the value 0.00, that is included in the matrix, acceptable?

……………………………………………..………Trust………………..…………….…Loyalty

Consumer innovativeness…………...0.08
…………………………………………………..…CI.900[-0.02;0.17]

Variety-seeking……………………..…...0.13……………………………..……0.09
…………………………………………………..…CI.900[0.03;0.22]………….. CI.900[0.00;0.17]

Thank you SmartPLS team

hussein
PLS Junior User
Posts: 3
Joined: Mon Jan 25, 2016 11:13 am
Real name and title: Mr. Hussein Issa

Re: Bootstrapping (5000 sub-sample)

Post by hussein » Mon Jan 25, 2016 11:25 am

Can someone help me understand why i can't copy and paste a bootstrapping report?
Chrome takes a long time and then stops responding!

Thank you,

Hussein

User avatar
Hengkov
PLS Super-Expert
Posts: 1619
Joined: Sun Apr 24, 2011 10:13 am
Real name and title: Hengky Latan
Location: AMQ, Indonesia
Contact:

Re: Bootstrapping (5000 sub-sample)

Post by Hengkov » Sat Jan 30, 2016 1:32 am

When the models are very complex and by using a large number of resample (e.g 5000), it will be difficult. Because the data you export amounted to very large. You may be able to use a smaller number of resample (e.g 2000) and further setting the export output options (this is available in the latest version SmartPLS).

Post Reply