|
|
Bootstrap standard errors |
|
Message/Author |
|
david posted on Monday, April 08, 2013 - 2:54 am
|
|
|
Hi I am trying to generate bootstrap standard errors for a CFA on binary indicators. I have used the commands: ANALYSIS: bootstrap=1000; output: cinterval (bootstrap); This produces the bootstrap CI's, but the se of the estimates remain largely unchanged when compared to analyses without bootstrap. Please can i confirm that the above commands are correct to produce bootstrap se and that the method to produce them is by taking the standard deviation of the estimates / loadings. thanks |
|
david posted on Monday, April 08, 2013 - 5:23 am
|
|
|
apologies, please ignore previous post - bs via R confirms |
|
Back to top |
|
|